There’s one issue, which is that of “Sherlocking” a third-party solution with a first-party implementation. Then there’s another, which is having an open enough system to support such innovations in the first place.

Something like Growl, or f.lux (mentioned down-thread) could never have come about if macOS had been as restrictive as iOS. I have little doubt that we’ve missed at least a few such innovations over past decade, especially on the iPad, due to this.

While the Mac will likely never, despite some tireless predictions, go fully locked-down, little things like the deprecation of kernel extensions will chip away at this from the Mac side as well. Market-driven innovation and platform control are a difficult balance, but they are ultimately a zero-sum game.

Well, yeah. Important things like tethering would never have made it into iOS if Apple hadn't been dragged there by the jailbreakers.

This has always been the domain of the carriers, not Apple.

I’m not from the US, could you explain this perspective to me?

You give me internet on device X. How is it up to you how I use it?

Same way it is up to me how you use books you bought (kindle) or songs you bought (any streaming crap nowadays).

If you're shitty enough person (good businessman) you can profit off of anything.

DRM on books is about the rights holder, not the person selling it. I can assure you that Amazon doesn’t actually give a shit what you do with your books outside of having an obligation to the rights holders.

Easy access to a huge library of books is one of the main selling points of the Kindle. DRM is good for Amazon.

Yeah, because they wouldn’t be able to offer it otherwise. If they could offer the same content without DRM at the same price, I’m certain they would do it. Apple did with the iTunes Store.

They wouldn’t, simply because their position as the dominant player in ePublishing gives them a massive incentive to lock-in users. Which is what their DRM does: you cannot move books to non-Amazon devices, so you’re stuck on Kindle forever. That this also satisfies publishers is something of a side-effect.

They did the opposite with music (they actually sold simple standard mp3s, unlike what you get from iTunes) because they were challengers, not incumbents, so the priority was to attract users.

iTunes has not sold music with DRM since 2009.

I assure you that they care much more about you buying books from them than about "lock-in" from you using their apps and buying their devices. Proof: their apps absolutely suck and their devices are not much better. The devices exist to get you to buy more books from them. It is absolutely not a side-effect that their DRM satisfies publishers.

In my experience in Canada from several years ago, the business teams in wireless telcos don't see it as providing you internet on device x. The ip connection on a device is a gateway towards a service, and the telecoms want to be providers of services, and the differentiator from competitors is having better and different services.

In my opinion the biggest fear of those business development teams is to be treated as or thought of just as providing internet on a device. As being just the pipe is not considered a lucrative market, and open to lots of competition where the race is to the bottom on price. This is also where you get lots of the anti-competitive behaviors, like not investing the effort to make a competitors service optimal within a network, as the carriers would rather launch their own service instead of making someone elses service they don't generate revenue off of work great.

To be fair, in the older gen access technologies, all internet access isn't equal. The access network was optimized for the usage patterns of a mobile browser, in the way it scheduled and idled the radio (I'm glossing over a large number of details). So connecting a computer via tethering that behaves differently and is far more chatty on the network, is actually a much costlier device to support. The current gen techs like LTE do operate much differently, to support chatty devices and apps. With a better optimized network and devices loaded with chatty apps, this difference is probably disappearing.

So if you put yourself in the shoes of a business dev person at a wireless carrier. They'll look at this and say, well we can have a $40 plan for just internet access. Or we can have a $35 plan for just mobile browsing, since that's more efficient and allows us to be more competitive when using the phones browser, and a $15 add on for tethering for those who want it and cost the company more. There are entire teams dedicated to just figuring this stuff out.

And this leads to some of the stuff I would get thrown from time to time, like the $60k bills (no longer allowed in canada, a rule was created to prevent surprise bills, but used to be a thing) when someone bought unlimited mobile browser for $7, and thought that meant unlimited usage for their tethered computer and started torrenting all day which was pay per use.

An unlimited $7 plan only works in the context of the way a user can use a mobile browser, and that's why the carriers think they can dictate how you use the internet connection. They have departments to tailor make plans that only work if they're able to dictate how the internet is used.

I'm not saying I agree or disagree with this model, I'm just trying to explain where I believe this perspective comes from based on my experience. I've been out of the industry several years now, and my opinions are my own.

This jives with my own experience. Great comment, thanks.

One group was trying to monetize services which were already free. Like make a list of data provided by stock mobile runtimes (Android, iOS), such as location data, and then create a developer portal for accessing that same data thru web services.

I'm having trouble remembering the details because I didn't even understand it at the time. My buddies were contractors, loved this kind of work, the money was good, and you could never fail. They thought there were helping me, get me some cheddar too.

I was briefly on the "mobile app services" team, it took me about 3 months to figure out what we were even doing. But there were lots and lots of meetings. Convened by empire builders and corporate climbers. Expensive suits. And lots of agile, kanban, scrum masters, velocity. To build something biggly impressive for which there would never be customers or revenue.

I assume there was at least one other group trying to "accomplish" the same thing.

The carriers would print money forever if they simply fired everyone unrelated to the actual network (cables and towers) and charged everyone a flat rate. And some overhead of the basic infrastructure services like provisioning numbers and 911.

No metering, no discounts, no custom phones, no carrier add-ons. The actual effort to create additional income streams was ridiculously wasteful.

It was crazy making. The actual features that would be awesome were never part of the conversation. For instance, I'd love to hear my voice mails on my laptop. I'm sure enterprise customers would love some tools for managing fleets of phones. Etc.

Maybe the carriers got high from the SMS profiteering, didn't realize that was a one-off bonanza, and have never recovered.

Much of what you said rings true with me and brings back chills of having a meeting to plan a meeting to discuss actually doing work, except for one piece.

> The carriers would print money forever if they simply fired everyone unrelated to the actual network (cables and towers) and charged everyone a flat rate. And some overhead of the basic infrastructure services like provisioning numbers and 911.

In my experience with these projects, I'm not sure it's safe to conclude that they add up to enough to significantly change the economics of the network. I know working on the projects feels like a huge waste of time and energy, but we're talking a million dollars here and there, when the cables, towers, and core people are spending 500 million plus per year. I'm sure it adds up, but all businesses have overhead, and I don't see it as a game changer.

As for charging a flat rate, I'm actually of the opposite perspective. I realize it's probably unpopular for this community, but for a long time I've been an advocate that mobile usage should be metered by use. The problem I have with the flat rates, is the users who use less always end up subsidizing the power users who use significantly more. Much of the network investment goes into supporting the top end users, but it's everyone else who has to pay for it. For wireline access this might be a bit different roi calculation, but for mobile wireless I think this is a real economics problem, and flat rate is not an incentive for a carrier to support or retain a power user who wants to use lots of data.

So personally I prefer a model that works more like electricity usage or filling a gas tank, where you pay for what you use, and you naturally get a feel for watching 4k video all day costs more. This of course needs the tools in place to understand where the usage is coming from, not to surprise anyone, etc, etc, so it's not a perfect model, but compared to years ago and the rates that could be offered, it seems atleast to me like a more natural model.

I'd be more okay with metered billing if:

- The fees weren't disproportionate with respect to any infrastructure investments. Why should my internet bill triple over the span of a few years while speeds decrease unless I call, wait on hold for a day, and ask to cancel service?

- There weren't an aspect of double dipping -- why are we being charged for peak bandwidth (with absolutely no guarantee of reaching those speeds) and _also_ being charged for exceeding bandwidth caps equivalent to a few dozen minutes of full use? Why is that "extra" bandwidth more expensive than buying several extra full internet plans?

- It didn't open the door to metering different content sources differently based on the ISP's monopolistic whims.

There are physical bandwidth limits, and those need to be allocated somehow, but the status quo isn't great.

Thanks for this detailed explanation. It’s really strange and I’m so glad this practice didn’t take hold where I’m from. Granted, mobile data is really expensive here so it probably wouldn’t make sense, anyways.

> This has always been the domain of the carriers, not Apple.

Maybe in the US. In Europe carriers never had any say about what functionality a mobile device offered.

All they provided was a SIM-card for that device.

Above what the sibling said regarding branding and pre-installed crap-ware, they still make a distinction between "on device" use and "modem use".

Source: read the fine print on an offer from my provider, Bouygues Telecom.

My regular plan has 40 GB, which I can use on the phone or tethered. This summer I was spending a lot of time at my parent's and their land internet connection was spotty. At the same time, Bouygues ran a campaign where I could get an extra 20 GB for next to nothing, so I was considering that. Read the fine print which basically said "the allowance doesn't apply to tethered use". I didn't care enough to challenge them on this so I passed.

Edit: Just checked the available options, this is still the case. Can't give a direct link for some reason, so here's a screenshot: https://imgur.com/a/rQgoz0m

Basically this is an option for "unlimited internet on the weekends". The originally hidden disclaimer says, among other things, "except for modem mode".

That's not true. Based in Germany, my mom still had, and I believe still has, a SIM in her phone that doesn't support tethering.

They did have a say, before iPhone and Android happened, devices have been branded, sim-locked, locked down etc by the carriers. I think they mostly stopped doing that in recent years.

I’ve been living in Norway since I was born back in the 70s.

We’ve never had carrier-branded phones. Not one. Only thing sold has been generic phones which accepts a generic SIM.

And that’s how the market is supposed to work. Free competition on devices. Free competition on service. Customers can combine as they like.

Granted you could buy carrier-locked phones rebated through a contract, but the carrier lock was time-limited and reversible and the phone was a generic, international model.

Carrier-branded phones was definitely not a EU-wide phenomenon.

If anything the introduction of the iPhone in Europe (launched using the very confusing US carrier-model) was what started pushing carriers into attempting to making new restrictions on how people were allowed to use their (formerly unrestricted) subscriptions.

So you got it pretty much 100% backwards.

You said "in Europe", when you must have meant "in Norway".

In the UK there were DEFINITELY carrier branded phones, tethering was disabled by many carriers, and you couldn't even use a regular SIM card in a non-phone device - you needed a "data sim".

I travelled around Europe for 2 years using local SIM cards - and also encountered carriers which disabled tethering.

There were even android apps specifically to work around these tethering restrictions, by making the phone act as a proxy.

Same for Austria, which interestingly was considered a "test-market" for international carriers, thus we always had the cheaper contracts and some novel business models quite some time before the rest of the EU (unfortunately not network generations or coverage in general).

The answer is money. Tethering was usually not allowed but you could buy in. You get phones for free, but only if you pay 40€+ a month for the next 2 years for something you actually don't need. E.g. some unlimited services (streaming) while your general data is capped.

I haven't been in the market for such contracts for quite a few years, this has changed a lot in recent times due to "contract-less" cheap providers gobbling up the marketshare. And these packages always disappeared over time and became standard. I don't think tethering is not allowed anywhere anymore.

This is how the free market should work. Providers tried to extract more money by forcing contracts, other providers swooped in with contractless plans and captured a significant market share.

I am pretty sure carrier-branded phones were a thing in Central Europe (Slovakia, Czechia, Austria, Germany etc) for many years. As the owners of carriers were also French (Orange) or Spanish (Telefonica/O2), I suspect they present there too. So no, not backwards at all.

They can put default APN configurations in the SIM card that (by default) would have different fares for mobile usage and tethered/hotspot usage. They cannot block you from changing the configuration, but I think in theory they could terminate the contract if you did so. In practice 99% people got used to buying both a "computer SIM" (which is typically sold with a cheap Huawei USB modem) and a "phone SIM", so the providers that do differentiate that way don't bother going after violations.

> In Europe carriers never had any say about what functionality a mobile device offered.

Not so, unfortunately. I tried putting a phone SIM into an iPad, and soon got a message from the mobile company saying the SIM card wasn’t intended for this use and would be disabled until it went back into a phone.

(This was in the UK a few years ago with a Three PAYG SIM).

Three is a special kind of nasty anyway.

They used to proxy all traffic and the only way to get out of their stupid slow proxy was to go to the shop with ID to get on an "18+" list.

They said it was because of some UK law but I was in Ireland. So not applicable. And other UK based providers like Vodafone didn't have this stupidity.

> In Europe carriers never had any say about what functionality a mobile device offered.

not really. a few years ago, before the eu roaming was made cheap, i bought a sim from wind operator in italy, and it was blocking my tethered traffic. i had to "fix" my ttl for it to work

Who do you think Ericsson and Nokia talked to when they planed their extensive and differentiated portfolio?

Oh, it would have made it, they just needed to figure out the best way to charge for it.

Was it ever Apple who charged for tethering? I always thought it was in the hands of the networks and their various plans. Today it seems like it's not even a feature anymore, just something you can do with your phone (or, at least for me it is: I'm on one of the cheapest cellphone plans in my country and tethering just works)

In the US tethering is fundamentally different (or at least the carriers want you to think that so they can upcharge it)

Yes, it is the networks that demand extra payment for tethering. That is still a thing in the US; you have to pay extra for a plan that includes tethering.

Surprising from a European perspective. I have used tethering since before phones had a HTTP/HTML browser (~ 2001, remember WAP?). The question has always been do you have packet data or not, what is your bandwidth and volume. Nothing else is the operator's business.

Not really. They still separate on-device use from modem use. At least Bouygues Telecom in France does.

See my other comment: https://news.ycombinator.com/item?id=25243710

hors modem, you are right. So France is different. I'm 100% sure that has never existed in Finland (well, the most common packages are unlimited anyway) and 90% sure it's not common in Germany either (they use throttling after you reach the limit)

So technically, how do they differentiate the traffic?

Most of the time, tethering is allowed in the base package. What happens afterwards depends. They can either cut you off or throttle. The only experience I have with this is Orange, and they throttle to the point where you basically have no connection (SSH is unusable).

Not sure though how they differentiate the traffic. There are "access points" set up which I presume are used for either connection. In my case they are the same but I seem to remember on an older phone they used to be different. It was also a different provider. I'm also not sure how they would detect that they're changed.

Screenshot of my setup: https://imgur.com/a/xH88Itp

Tethering and hotspot traffic is still marked with a different APN type ("dun").

Does APN type go out over the network?

I thought that's an Android internal thing to decide which of several APNs to use.

Also the first 2 references that Google returns say dun is obsolete and not in use with today's devices. I believe that belongs to the ATD *99# era

(Havent't worked in the field for 10+ years, could be wrong.)

No it doesn't go out, but the APN itself does go out on the network and can be used to apply different billing to tethered vs mobile browser data usage. https://news.ycombinator.com/item?id=25244052 for more info.

The networks demand it, but Apple provided them the ability to demand it.

Actually the same thing did happen on iOS with jailbreak tweaks being copied by Apple devs. (E.g. wifi sync, quick reply, predictive keyboard...).

Admittedly the pool of developers was restricted to those who were able and cared to jailbreak...

It's interesting to think about the fact that, now that Macs are starting to be differentiated more by hardware (again, due to Apple Silicon) than software, Apple might actually allow more tweaking/low-level access to do interesting stuff.

Performance of A14X is almost the same as M1, but it only became a sensation when the CPU was allowed to run macOS.

Think about it: Apple already had an awesome CPU, but hardly anybody noticed, because iPad OS is so limited.

Not true, there is no A14X (yet). Closest thing would be the A12Z in the 2020 iPad Pros which are impressive but ~25% slower than the M1.

> These are real, material examples of the benefit brought by developers being on your side.

I think it actually goes a little further. Growl is (was) a fantastic notifications system, but it wouldn't have succeeded without the healthy ecosystem of third-party programs that were willing to use it. When I installed it, long long ago, everything I wanted notifications from suddenly supported them. If the developers hadn't been talking to each other, it wouldn't have worked.

That’s really the key to our success with it, but it’s more. When a user found out about Growl and liked it, then they would request one or two app devs to add it. Those app devs did, the. Their users got it. Rinse and repeat. So it’s both developers and users who made it possible.

How should Apple have handled this? Maybe insisted on acquiring the Growl company? I don't think Apple should have just avoided building Notification Center, since that's a big net benefit for everyone.

If an acquisition is rejected/infeasible/not applicable/etc, then I'm not clear on the right thing to do. Acquisition might have been possible with Growl, but for some other cases there's not even a company to acquire. Have any other big platforms done this well?

(Apple's acquisition of Workflow which became Shortcuts seems like a case where they did this well)

I don't think OPs frustration was with 'sherlocking' Growl, but with Apple's stance on iOS and the increasing 'lockedownedness' of MacOS.

The MacOS community, and I think many MacOS developers accept that 'sherlocking' is a thing and I see it as something that should be a point of pride for these developers: "we built something so good that Apple decided to rip it off" one oft cited Steve Jobs (through Picaso) quote of course being "good artists copy, great artists steal". But I do understand developers who are frustrated by this happening to their apps, and don't begrudge them for it, especially when it is their source of income.

Acquisition by Apple would have been an option if they had contacted us.

They could license the IP to forestall copyright/patent challenges and motivate creative devs, similar to security bug bounties.

I would guess that there is some legal problem with even referencing original app as an inspiration. Paying bounty for it would be even worse.

If Apple were able to copy some app and in return mention your name as a co-author of some sort/tip you - that would probably be a life changing experience for most developers out there.

> If Apple were able to copy some app and in return mention your name as a co-author of some sort/tip you - that would probably be a life changing experience for most developers out there.

Losing your income stream certainly is life-changing.

How would Apple admitting they did it change your life for the better?

When you say "tip you", are you thinking "compensate you for all future lost sales and throw in an extra million or two"?

Do people get tweak-y with their systems as much any more?

I used to get extremely into customizing Mac OS (as in, Classic Mac OS), and early versions of OS X and iOS during the jailbreak salad days.

Now, not so much. I tend to run closer to stock, and not deal with the system constantly changing and deprecating my tweaks. Is that because I am old, or is it less common now? (Not a rhetorical question, I really don’t know.)

Go look at any iOS jailbreak community, it's chock full of people with nothing better to do than endlessly tweak their clock screens or whatever.

It’s us. I was the same, now when I have a fresh Mac (like at work) I like to grab my dot files, my ssh keys, and remove a few items from the dock and I’m good to go. I don’t even do the last that much since I mostly launch from Spotlight.

It's similar to people building their own computer rigs. I built my own computers when I was younger and broke. It was cheaper to buy the parts and do the build yourself than it was to buy an off the shelf solution. It's also a great way to learn about the computer. As I got older, I would still lean toward building computers. However, today, I have way more responsibilities to be building the hardware, and my time is way too valuable to care about doing that myself. I just need the damn thing to work, and when it doesn't, there's a person to contact about resolving the situation. Between work, family, hobbies, etc, I have chosen to no longer care about these customizations/tweaks. There's really nothing different today for me to learn about a computer by doing the build myself: cpu, ram, bus, i/o controllers, etc. The core fundamentals are the same, just tech and speed has changed. My current devices have stock backgrounds. Rarely do I see my desktop as I have too many windows open doing actual work.

TL;DR I too care much less about this, but understand why others do. I encourage them to keep doing it even if I don't personally spend time with it.

I still build my own PC's even though on a similar cost/benefit basis it would make sense not to perhaps.

Thing is I build PC's for both gaming and software development and frankly for that goal the delta between build it yourself and someone else building just on price is pretty large but there is also the fact that if I build it myself I know exactly which parts are in there.

There are a lot of people customizing their OS X experiencing (including custom window managers). See here for some examples:

https://www.reddit.com/r/unixporn/search?q=mac&restrict_sr=o...

I still have yet to find a satisfactory replacement for Quicksilver. Yes, Alfred and all those exist. They are mere shades by comparison.

I feel like Launchbar, which I believe predates Quicksilver, is pretty great. It’s true Quicksilver really perfected that motif of doing custom search operators and file manipulation, but Launchbar is really great.

I should probably give Launchbar another shot at some point; recommendation appreciated. My expectation is merely that it’d be nicer than Alfred, but that’s enough.

Quicksilver was just updated in June.

https://qsapp.com/changelog.php

It’s fair that the core app still compiles and has some maintenance, but there hasn’t been work done on the plugins and system integration in about a decade. What’s left makes a nice but slowly decaying launcher. The magical workflows of being able to navigate swiftly between app context menus, system services, and web services without a mouse or even leaving QS...those are but memories.

QS circa 2008ish spoiled me for all launchers since.

> QS circa 2008ish spoiled me for all launchers since.

Same. I (or somebody else) really should port Quicksilver to Linux; it's the day-to-day utility I've missed the most.

Just curious..what is it you find lacking about Alfred? I swapped a few years ago when QS was having some wobbles with releases keeping up with Mac OS releases. I forget what the exact issues were at the time. It took about 3 months to find and replace my QS workflow but now I love it.

It turns out that this is now hard to answer! I tried finding the original articles and screencasts I’d used to get started, and they’re all on dead domains and Google Video. The tl;dr version is Proxy Objects And Their Implications. QS was (and remains) a serviceable launcher, but it shone most as a way to construct functional sentences, similar to UNIX pipes but for the MacOS GUI. Being able to do trigger->Current App Menu Contents->(type name of menu item) was already rad. “Select all these files, compress them into a single archive, then attach them to an email with the following subject” from your launcher was doable and even straightforward. “Capture text/current clipboard and apply a preset template for this keybinding and append to my Capture file” was essentially the first pass for how people did GTD capture on Mac.

Granted, a lot of the is now doable with specialized apps, but to no longer have so flexible and versatile a tool makes me sad.

I can’t build demos at the moment, as my Mac has been in the shop with Apple for almost a month now with precious little communication—but that’s another rant entirely.

Raycaster is the only thing is looking better than alfred for me right now

I was unable to find this with a bit of googling; do you have a link?

https://raycast.com I guess

Yes that's correct, sorry for the misspelling on original comment, can't edit now.

It's a pretty young company / product but I see great awesome first party integrations instead of unreliable hacky 3rd party scripts which was the default on alfred/packal median of plugins tbh. (As much as I loved 'em and toying with them)

Raycast integrations with for example another awesome tool (Linear.app for PM) Github Issues or Jira if you or your company feel the necessity to put yourself thorugh such painfully useless endeavour lol)

PS: When looking for alfred scripts the other day for a co-worker which did a clean install after recommending it to swap spotlight, packal returned pretty ugly errors when running any search keyword on the site, doesn't seem that well maintained either... so

Hopefully Linux will be the operating system to benefit from increased development. I simply don't understand why developers keep investing their time on platforms that are openly hostile to them.

These posts were written decades ago:

http://www.paulgraham.com/road.html

https://www.joelonsoftware.com/2004/06/13/how-microsoft-lost...

> If you want to write desktop software now you do it on Microsoft's terms, calling their APIs and working around their buggy OS.

> And if you manage to write something that takes off, you may find that you were merely doing market research for Microsoft.

It keeps happening over and over again!

> I simply don't understand why developers keep investing their time on platforms that are openly hostile to them.

It's because the users are there. It's similar to things like YouTube. Every YouTuber complains about YouTube... but there is no other place where their random video is going to be recommended to a million strangers. (Twitch is similar.) So, they put up with it.

It boils down to what problem you want to solve. If you can figure out how to convince Mac users to switch to Linux, then you can be successful in your approach of "ditch Apple for being evil" or whatever. If you can't, then you have to find a new line of work (there is plenty of software engineering to be done that never touches an Apple product), or you have to put up with the poor developer experience.

It's also unlikely to be sunshine and roses on the other side of the fence. For everything that's bad about platform X, platform Y probably has just as many annoyances. If you're looking for perfection, you're going to have to remake the world in your image from scratch. That's a lot of work!

It's hard to make money building for an OS like Linux whose users expect everything to be free and open source.

If you want to sell software for money, Linux app development is not the right business to be in.

It’s because no one (mainstream) uses Linux. Developers make these projects and still want lots of users and people to find use in the products.

Developing for Linux leaves you with a smaller range of users than Windows and Mac.

Smaller group and often less willing to put out cash for it (besides the few who would be willing to pay substantially more to get great applications on Linux).

This was the belief for a while, but it seems that isn't totally true.

https://www.compoundtheory.com/some-arguments-against-linux-...

https://blog.hiri.com/a-year-on-our-experience-launching-a-p...

"Pricing wise, we haven’t noticed anything that distinguishes Linux users from everyone else. They are no more cost conscious than Mac / Windows users. They are definitely willing to pay for software."

As an on-again/off-again linux user, I have paid for software or donated to many projects and it seems others are willing to do the same, however it is still a much much smaller group.

I know many developers working on Linux laptops (web backends.) I do since 2009. Would I develop some GUI tool with the aim to sell it? No chance. Gnome or KDE, apt, yum, flatpack, snap, etc? No thank you. I take what's available and that's it. Ubuntu has proven for 11 years that's more than enough to give me a desktop I like (currently a heavily customized Gnome Shell) and pay my bills.

Don’t forget f.lux, now Night Shift.

I still use f.lux because I find Night Shift doesn't get red enough for me.

On MacOS I found serious performance issues with f.lux and now I happily use redshift[1] without ever thinking about it.

1. http://jonls.dk/redshift/

f.lux works by simply setting a new ICC profile - right? That’s often supported directly in hardware in the GPU’s scanout system (so it doesn’t appear in your framebuffer). So why would that cause poor performance?

yea never had any performance related problems on the visual end of things with flux (typing this with it on). Although flux is getting pretty outdated on Big Sur, so there are some hiccups and lockup bugs that could be interpreted as "performance issues".

I dunno maybe just an older version. Maybe the fade or the menu.

f.lux had major problems. There was some sort of weird issue that seemed like they were not unpremultiplying colors before shifting the colors, so you'd get these weird blobs of color that either didn't have the color shift applied or were just the wrong color in the middle of videos. It was quite annoying.

> you might find another OS starts to benefit

I actually really truly hope this happens. I want to see the same sort of love macOS gets from developers, showered on some/any open source OS (e.g. on some Linux/FreeBSD/etc distro).

I've been using Linux for nearly two decades, since I was 12 years old. (I still remember the excitement of installing Linux dual boot on my parents' PC years ago.)

I've been wishing and waiting for the age of the Linux desktop to come. It feels like it's so close, yet so far away.

(I've also been contemplating the benefit of me sinking the time into creating yet-another distro of my own -- one that's a lot different, built upon the features from NixOS and GoboLinux -- a distro that can hopefully be a truly compelling OS to wide range of folks...)

For the record I'd love this too, I just don't see it happening anytime soon.

I've been thinking about this recently I'm not not sure the (vocal) Linux community would accept what it might take.

* Developers want to work on projects that interest them and provide a benefit to others.

* However, developers also want to make a good living so they need an audience willing to pay money and make it worth the time it takes to polish something to a decent finish.

* Some developers would prefer to keep their code closed-source.

(Again the vocal) Linux community all to often comes across as everything should be not only be free open-source also free to buy - it's almost a dirty word if you charge for software.

Additionally on the Apple-side of things:

* There's a culture of what constitutes a good app, it drives a certain perfectionism to the final polish that you rarely see in linux desktop apps. Personally I've not seen a huge amount of apps on linux that cater for different user audiences. As technical aware users we vastly over-estimate the amount of technical knowledge and patience an average user has to figure something out.

* Apple is now offering an audience from a multitude of devices. You can build your app for a watch, phone, tablet or desktop. e.g. if someone buys your app on the iPhone they are more likely to be interested in your apps for other devices so there's more opportunity to cross-sell.

Ubuntu is probably the closest I see to being able to set some proper direction here. But I've yet to see them double-down and really set their mind to it, they seem to set a direction hold for while then back-down and go another direction. From the outside, it seems like anytime they've really tried to do something different or _the horror_ make some money it seems to just rile up the vocal linux community.

> Apple folks: take note. These are real, material examples of the benefit brought by developers being on your side. Shun them as you have in recent years and you might find another OS starts to benefit from their weekend projects and innovative ideas.

It doesn't matter when some developers condone the forces that take advantage of them because of incentives(stockholders) or general apathy.

> Apple folks: take note. These are real, material examples of the benefit brought by developers being on your side. Shun them as you have in recent years and you might find another OS starts to benefit from their weekend projects and innovative ideas.

I've recently come to a shift on mentality and believe that such sentiments are meaningless. It's almost always the most parasitic and immoral of the players that succeed and continue to succeed. The worse they can treat the other parties, generally the better off they are. It's like a deer telling a lion it should consider eating more grass.

Yeah it sucks cause we're generally in the camp that is being taken advantage of. But due to the forces of capitalism and human nature, you can practically mathematically prove that your words will not be heeded, and to great great profit.

Was Spaces an app before it was a feature? Currently I'm using third party software to pretend I have Spaces instead of the awful thing that replaced it.

What are you using our of curiosity? I used to use the amazing TotalSpaces but with recent versions of OSx it was no longer able to work. Now I've just gotten used to OSx's shite version but I really miss using a grid

I'm using TotalSpaces on Catalina. The web site says it works on Big Sur but not on Apple Silicon. A developer posted that a future release will be compatible with Apple Silicon: https://discuss.binaryage.com/t/future-support-of-apple-sili...

Do you still have to disable the security setting? Can't recall the name. But that's interesting I'll look back into it.

Apple bought Cover Flow. And then retired the concept for the most part since it really wasn't that good of an idea.

How do you send a notification now without Growl? Each time I try to google for a tutorial, they all say “just use growl.” Ultimately I gave up and installed growl, and it works quite nicely.

I’d just like to pop up a notification programmatically via a bash script. :)

You can make an AppleScript thing and call it from bash. I think this is the sort of documentation I used at one point: https://developer.apple.com/library/archive/documentation/La...

AppleScript does work well for this. It might also be nice for somebody to write a little UNIX-y command-line app that just allows you to pop up a notification, and potentially get a response to it. Maybe I will...

Thanks!

I'm using Pushover (https://pushover.net/) to send notifications to my Apple devices. To make use of it easier on command line I wrote a simple Python script which I've found to be very useful for things like allowing Transmission to send notifications when a download is finished or in scripts that do backup jobs which take a long time, letting me know when my Mac has booted, etc... Reply to this comment if you're interested in my script (note that the pushover site has good examples of how to call their API).

terminal-notifier does this.

https://github.com/julienXX/terminal-notifier

Pushover. Dead easy to use in a bash script etc with curl or wget. It also has an email gateway.

Similar things happened with most of the popular jailbreak utilities for iPhones. Custom backgrounds, app icons, ... really all of the successful modifications were eventually pulled into iOS.

Right. iOS was very feature-anaemic for its first 6-ish years. I switched to the iPhone in 2011 after being a faithful PocketPC/Windows Mobile user for more than a decade - but I had to jailbreak my iPhones to get the system tweaks from Cydia that I felt I really needed - things like a Today screen, AdBlock, raw file system access for exchanging files with my PC (and for getting my data out of oppressively siloed applications, etc.)

Since iOS 8 the system has had most of that functionality baked-in or is officially supported by Apple’s APIs for third-party devs. The only real things I feel I’m missing out right now on iOS 14 is raw FS access and easy sideloading.

I don't need raw file system access nor do I think most do, but using the phone as a USB drive is something I've always wanted and had many situations where it would have been useful. Apple could enable iPhone and iPad to be encrypted general storage devices.

> easy sideloading

Workaround available. [1]

[1] https://altstore.io/

They definitely had this functionality since the iPod (being a general usb storage). I haven’t tried it since starting to use web storage (eg google drive) but used this feature in high school with my original iPhone and iPhone 3g.

> I haven’t tried it since starting to use web storage (eg google drive) but used this feature in high school with my original iPhone and iPhone 3g.

Unless you jailbroke your iPhone 2G and 3G that's an impossibility: the iPhone has never officially supported USB Mass Storage mode: it has always only ever officially supported MTP/PTP (and DFU) over USB.

Shouldn't that be anticipated though? The developers at Apple (or any developer in general) cannot possibly think of every little thing that user might want to do with their device. That's why nothing stays at v1.0 for very long. Users report bugs, and even make requests (however that might look). Any developer not looking to incorporate these requests will see their product wither and die.

Sure, sometimes a 3rd party comes along and makes a great product that once it is used, it feels like it is just something that should have always been there. Apple being Apple, they are going to want full control, so if they can't acquire the tool to do what they want, you know they will develop it internally. Every tech company does this. FB/Snap/Insta/etc have all borrowed/stolen/re-implemented.

You could just as easily say it was driven by Android competitors.

There aren't really that many ways to do notifications.

If you want something that pops up on a large screen but isn't overly intrustive, a rectangle with text in it in the corner next to a menu icon to control them is what you're going to come up with.

And if you're Apple it'll be a rounded rectangle.

And Apple absolutely did not rip off Growl "pixel for pixel", the visual styles are totally different. I don't know where you even got that from?

Not only that, but Apple already had a similar feature in MacOS 8 and 9 that Growl was reproducing in OS X. It seems like Apple just hadn't gotten around to putting it in OS X yet because other things were more important. Then eventually they had time and added it.

I never used macOS until OS X and most definitely did not copy it. Nothing about Growl was a copy of that.

Growl was inspired from being project manager on Adium and worked with the people making colloquy. The devs on colloquy were working on notifications and so we’re we and I thought it was smarter to make it a separate tool.

Hey Chris, for what it's worth, your software was quite a lifesaver for me in my earlier OS X days back in college. Your software had great impact on my workflow and I'll never forget it. :)

Happy to hear it. We’re totally didn’t know what we were doing really, just making something we wanted.

It just illustrates how natural the concept is. I don’t mean that in a bad way; it’s such an elegant solution and I would not be surprised to see other implementations before MacOS 9.

Anyway, Growl did much more than that, and I don’t think there is any doubt Apple’s current implementation owes a lot to your design.

Yea I just wanted to clear up the inspiration bit. I’m sure if we hadn’t done it someone else would have.

Fair enough. Thanks for the clarification.

Not very similar. If you wanted a message, the OS put up a modal dialogue blocking everything else. The whole point of Growl was that it didn't interrupt you, and you could have its messages go away on their own.

There were these yellow messages that popped up in the upper right-hand corner in MacOS 9. Those were just non-blocking notifications.

Exactly. They look a lot like Growl and the macOS X notifications and were not modal.

Wait wat? I’m not disputing this but I came up on MacOS (it wasn’t even named that at the time, I believe I got started on System 6) and I never saw notifications like that.

Around Mac OS 9, they changed some types of errors from modal dialogs to floating palettes in the corner of the screen.

I’ve never seen these before. Sounds interesting but not the inspiration for Growl by any means. :)

All great design innovations seem dead obvious in hindsight. But they're not.

Sometimes the design space is so limited though (I'm talking in general here), multiple people will inevitably arrive at similar designs. I've felt this myself with mobile design which can be really constrictive. Once you enumerate obvious interface variations, weigh up pros/cons, discard the obviously bad ones, iterate a little etc. you can come up with something you think is unique but has been done many times before.

I was designing a mobile paint app for example and had to make choices like how you bring up the colour picker, move layers, change tools, preview brush settings etc. There's only so many ways to do certain things on a tiny screen.

Huh? Notifications that pop up in the corner aren't a "great design innovation". They are an obvious one.

Just because there are some non-obvious great design innovations out there, doesn't mean every little interface component is one. Some things are just the logical solution to a problem.

Oh? In your estimation, how many years passed between Growl being technically possible and it actually existing?

I was using NeXT OS, which is what later became OS X, circa 1990. Growl apparently launched in 2003. So that's a minimum of 13 years that something "obvious" was missed. And it's probably more fair to count from the mid-1980s, when GUIs first started becoming popular. That doesn't sound obvious to me.

Most things are obvious in retrospect. But it's a mistake to confuse your after-the-fact perspective for what was going on at the time. (For those who are interested, Dekker's "Field Guide to Understanding 'Human Error'" is a great look at how subtle and dangerous that confusion can be.)

I don’t think it’s that easy.

Timing is a factor, too. For such a thing to be successful, you also need a OS/hardware combo that can draw the notification without slowing down using the main window (rules out early Mac OS) and users who think the added distraction of notifications is worth it. I would that added distraction goes down with screen size. I doubt it would have been a success on 640×480 displays, for example.

Also, Apple had something similar in 1997 or 1999, with Mac OS 8 or 9.

Fair point. 1984 is probably too early. But the NeXT, which was launched in 1988, meets the technical criteria: big screen, fast display rendering, background processes. It had ongoing notification via the dock. And as somebody who used them then, background notifications would have been way better than interruptive modal dialogs seizing the foreground, which were the common UI choice.

Pop-up notifications from background processes only really became necessary once people started having a lot of internet-connected programs that needed to notify you of incoming things. Back when the only thing was your e-mail inbox, you didn't need a separate notification service.

Obviously then it was then extended for things like a long process completing, etc.

But back in 1990 or 1995, there's wasn't a need for something like Growl. Your e-mail inbox and the occasional beep and modal dialog did the job just fine.

To counter your "most things are obvious in retrospect" philosophy, you might be interested in the "multiple discovery" viewpoint [1] which says precisely the opposite -- that, extended to design, essentially says that the need for a solution becomes obvious to people at about the same time, and that people will solve it in similar ways because they're facing the same constraints.

[1] https://en.wikipedia.org/wiki/Multiple_discovery

Modal dialogs were always terrible. They were just easy to code, and fit the common programmer mental model of code first, not user first. The NeXT folks clearly recognized that background activity was a concern given how they were using the dock to give app-specific notification. But design-wise it was a cul de sac because it required permanent screen real estate for anything that might matter.

I also don't think multiple discovery is much of a counter to "most things are obvious in retrospect". Yes, design problems can get solved in similar ways by different people. But obvious-in-retrospect thinking is a cognitive error, where we presume that what's obvious to us is obvious to people in different times and conditions.

Growl did more than pop into a corner. :)

Oh absolutely. :) But I assumed the accusation that Apple copied Growl was mainly that.

Growl had so much configurability, so many options, it's not like Apple implemented really any of that!

Notifications could also be shown in the menu bar, with marquee scrolling. No need for them to pop up in a corner.

Marquee scrolling is virtually always a terrible UX pattern.

Also oftentime there is zero extra space in the menu bar.

Alerts and notifications have followed a pop-up pattern since basically forever in computing.

About the space: I was thinking the text could replace the icons on the right hand side for a short while (until the user has had the time to read it).

Emacs displays little notes in the echo area at the bottom of the screen, though those are always (?) triggered by user actions and do not come from background activity. Which also qualifies as "since basically forever", I would say.

> They are right to say this, as the current Notifications system in OS X is ripped nearly pixel-for-pixel from Growl's implementation a decade ago.

I _think_ when Growl came out it was an open-source implementation of a notification UX that Apple had already demoed, either as a prototype or in some first-party apps, but it's just a vague memory. Does this ring a bell for anyone?

Nope, I came up with it and that didn’t happen.

I think I was thinking of iChat. If someone started a conversation with you while it was in the background, it would display a small pop-up window in the corner of your screen, which you could click to start the chat. It appears in a screenshot here (from 2002): https://arstechnica.com/gadgets/2002/09/macosx-10-2/12/

So the basic design of the notification UI (small pop-up window, short and long, icon on the left, click to respond) was already in iChat. Other chat clients for Mac (Adium, Colloquy) were trying to imitate it, as I recall (and ended up improving on it).

This is also interesting on Growl's early history: https://web.archive.org/web/20060115015145/http://www.drunke...

Who is shunning developers on the Mac? If anything I think most Mac users lament the dramatic shift toward cross platform/Electron apps.

I hate those apps really. I use them because I need to but they aren’t fun to use.

And locking the OS behind a gilded cage and not letting developers tinker. I’m done with MacOS as of Big Sur. M1 is massively intriguing but I am done with macs for a while. Over half of my companies 2018 macs had battery bloat and broken keyboards. Bloat breaks the keyboard. Half! It’s a real shame. I can’t abide by it anymore, as a developer I feel shunned by the OS now.

They're not stopping anyone from tinkering.

The last few weeks have been non-stop rave reviews of the new Mac. I doubt they’re worried.

I’m not sure I’m following this comment, the vast majority of those Macs are being purchased to run macOS, and macOS is immeasurably better because of developers making things like Growl.

Note that Apple has dropped hints that they’d like users to move from macOS to iOS (“what’s a computer?”). Third party developers can’t build anything like Growl on iOS. There are a lot of reasons I don’t think Apple will ever be able to replace macOS with a closed system like iOS, but at least one of them is simply that a system as closed as iOS will inherently have a low ceiling for innovation (not enough room to innovate on the platform) therefore this innovation will be channeled into other platforms.

In other words, I think developers building features like Growl for macOS is highly relevant to the success of M1 Macs.

Is this actually true? Notifications were in iOS way before OSX, and it seems like that is the actual implementation that eventually made its way into Mountain Lion.

I mean, Toast notifications weren't even invented by Growl. It was nice software but this seems like hyperbole lol

FWIW, Growl predates iOS

Edit: also, IIRC Growl was originally called "Notification Center", which is the name that Apple later used for their implementation of similar features

Double edit: I should've RTFA which mentions my fun fact

IIRC, Growl was created because a similar feature of Mac OS didn’t make it into the first release of Mac OS X (sorry, can’t find pictures or documentation online)

If so, I think its developers can’t be completely unhappy about the end result. Also, their code is (3-clause) BSD licensed. If the product lives on in spirit, does it really matter where the code came from?

I never used Mac OS until OS X because I wanted a terminal with Unix userland and also ms office. I don’t know what feature you’re talking about and it was not a consideration when coming up with Growl.

You’re right, but Growl did much more than the notifications in classic MacOS. Also, Apple took their sweet time to get their notification system for OS X.

I first got into the Mac eco system in the mid 2000s, so I don’t know the entire history of notifications there, but I remember that Growl was pretty much the defacto Mac standard by the late 2000s.

Growl and a particular packaging software that provides auto updates had basically become a prerequisite for Mac apps at a certain point.

The auto updates software you're thinking of is Sparkle, and it's still going strong. Still by far the most common way for Mac developers to do automatic updates for apps distributed outside the Mac App Store. https://sparkle-project.org

Yes, but notification center, added in Mountain Lion, was an exact clone of the iOS 5 notification center. Which at the time was argued to be stolen(!) from Android's notification center design. So did Android steal Growls implementation, as well? That is what is being implied.

The update framework is Sparkle and is still widely used. You’re quite right that these two frameworks made important chunks of app functionality Just Work.

Growl goes way back, this is an interview from 2006:

https://www.osnews.com/story/15442/interview-with-chris-fors...

Good lord I forgot about that interview. I was so young haha.

Growl started around 2003 really

I’d just like to thank you for your work on Growl, Perian, and Adium. Those were fantastic, provided great functionality and were rock solid. All of them were invaluable. They are always associated with the good time I’ve had tinkering with my computers as a student.

> Toast notifications weren't even invented by Growl

Does MSN Messenger get the credit for that?

Windows 2000 according to Wikipedia

MSN Messenger predates Windows 2000 (July 1999 vs February 2000 respectively).

It’s possible Apple independently arrived at a design for how to present iOS notifications that just happened to resemble growl in every way... but it’s not likely. Growl was so pervasive before macOS notifications that it’s inconceivable to me that developers at Apple weren’t presenting it as a proven model.

Uh huh. This is ignoring that notifications were present in webOS, Android, Windows 7, and so on. Or were those able to independently arrive at the exact same design of a toast popup, and Apple uniquely owes the provenance of their toast popup to Growl?

Growl existed before webOS, Android, and Windows 7.

Windows have system tray balloon tips since Windows 2000. Or even earlier

What is your point? Is Growl responsible for all notification implementations post 2003?

Wow, everyone is so fucking combative and contrarian the last few days. Chill out, that's not what I was saying.

It seems to be exactly what you're saying?

I was only saying that it seems pretty likely Apple took inspiration from Growl. Not that they didn’t take inspiration from other prior art, or that Growl didn’t either. They’ve taken similar inspiration from other things on their own platform even where there were multiple inspiration sources available. Spotlight took design direction from Quicksilver even though Enso existed. From Mac OS 8 (or was it 9?) through most of the OS X/macOS span, it took inspiration or cues from WindowShade even though a zillion other minimization styles existed. They literally bought CoverFlow even though a ton of other presentational equivalents existed. It’s not that these ideas don’t flow around everywhere. I was just observing that they likely observed what was already at home on their platform.