> 2005! The future is here! You have just spent $129 for the newest release of Mac OS X: Tiger.
For me it was more like “you have just torrented the Golden Master DVD image and restored it on to your bootable Firewire iPod because you only have a CD-RW drive and nobody has released rips of the six-CD version yet” ;)
The bootable disk function of iPods was really something else. It saved me multiple times through the years, at one point even functioning as a primary boot drive when the HD in my sunflower iMac G4 gave up the ghost.
I wish modern smartphones had a similar capability. I know Android phones let you copy files to them via MTP, but that’s not even a fraction as good as the portable HD function of iPods was.
I think it could be a killer feature for iPhone + Mac
Here is a secure external disk, which you can plug into your mac, it will be secure even when plugged into your mac, even the data over the thunderbolt bus could be encrypted in transit, its all technically possible to do.
It never shipped obviously, but back in the day there were a lot of rumors of Apple developing a feature that let you take your home folder and apps with you on your iPod, with any Mac you plug it into making your user account available without even rebooting. So it seems that they saw that use case and almost acted on it.
Unfortunately the oldest snapshot in Wayback is October 12th 2003 with the blurb already removed. Apple filed a patent application for it in 2002 which was assigned in 2006. It expires this year!
Surprisingly, there have been a few vapourware announcements by Apple over the years. Another was the “Restart to Windows” item under the Apple menu when Boot Camp was first announced.
Would love to have been a fly on the wall after the keynote where Steve chewed out the engineer that left that in.
MTP sucks from the beginning to even now. It's too slow for small files and often stuck. I wish any improvement come in this area (file based wired transfer protocol).
I've used it successfully in the past to serve different Linux ISOs on the fly. It relied on certain kernel features / permissions to act as a USB mass storage device.
That’s true, I remember being a bit confused when I found that Windows was extremely picky about what it booted from, with weird hacks being required to boot a Win2K or XP installer from anything other than an optical disc. Both OS 9 and OS X were pretty lax about that.
Yeah I booted from USB HDD for a while in the XP days but it was necessary to edit the registry to make sure all USB related drivers would load at an earlier stage during boot.
And installing any updates would usually revert this, leading to an unbootable system :/
Not sure if this still works like that but it was a total PITA.
I paid for alternate OS X versions (and then jumped on the snow leopard wagon immediately due to the $29? price).
That meant shelling out $129 after about 3 years (2 year lifecycles, and I think I bought my first mac almost a year into the then OS lifecycle).
This provided me with excellent stability and a very reasonable price. I'd save up a few hundred $s, and then upgrade the software and hardware at around the same time, so I also added additional RAM and moved the HDD to the CD-RW and inserted an SSD instead of the HDD (I am 100% sure I made these changes for my macbook, but I'm not sure if I made them, or if they were even possible, for the iBook I owned before).
That was almost a decade+ of highly stable, highly effective, and almost cutting edge of computing that I haven't even come close to replicating in the almost decade since, despite earning real money.
I did switch to Linux for my personal computing a few months ago, and I have hopes that this may allow me to do so, once I really set something up once I get to my Mar-Apr spring cleaning. Linux is giving me that Mac feeling for the first time in a long time, although the major challenge here appears to be restraining oneself. It seems so easy to get lost trying to distro hop constantly, or try a new terminal for marginal benefits, etc. The new shiny in Linux shines very bright, and restraint seems to be the core challenge required to have a stable, outcome focused computing experience with Linux.
I had a disc i bought at the end of line to fix a mac of a friend, some decade ago! and when refurbishing old macs to Linux become very handy, as before i could install linux sometimes i needed eEFInd.
I don't know if it's nostalgia but somehow this interface looks more lively and friendly to me than modern MacOS which feels flatter and has less personality (to me personally).
There's definitely some ugly bits from that era, but I miss...color.
My favorite example are the icons in the iTunes sidebar, which used to be distinctly colored (and therefore quickly and easily distinguishable!) and then became a dull grey. It looks incredibly bland and it's harder to use.
Now it's split into multiple apps with monochromatic sidebars, which is slightly less boring but no more usable.
I had some firsthand experience of how many people were bothered by the grey iTunes icon change after I released a little hack to restore them. The demand blew through a month of my web hosting bandwidth in a couple of hours, and people kept sending me emails about it for years! Comment from a previous discussion:
Tiger remains to this day my favourite look for macOS. It's not that I don't like the modern look, but there's something incredibly friendly and inviting about the 10.4 era Aqua interface. I wish you could still skin macOS because I'd totally run a Tiger desktop on modern macOS if I could. Best I can get is a brushed metal theme for Firefox to mimic old-school Safari!
Tiger: Great new UI. Spotlight. Dashboard.
Leopard: Quick Look. Time Machine. I hated it from a functional perspective, but wow it looked amazing.
Snow Leopard: They just cleaned up everything.
Early versions of Snow Leopard did have a bug that could delete all the data on your hard drive. :)
10.6.8 was pretty good, but I've had similar stability on Mountain Lion (10.8) and Mavericks (10.9). I think Snow Leopard has developed a legacy partly because Lion (10.7) was pretty bad, and partly because Snow Leopard was the last release to support Rosetta.
Then again, Yosemite (10.10) was also really bad, and I seem to be the only one who thinks Mavericks was the actual pinnacle. *Shrug*
Snow Leopard was great to work on. When we were told the next release of OSX would focus on stability and performance, morale went way up. Having a massive backlog of bugs felt like an unseen energy vampire, sucking away the will to live. You just knew some user out there is hating life because you don't have time to address their issue. The Radar issue count for Finder was in the tens of thousands alone and it could be argued each bug was worthy of being fixed.
Do not fear--it's only a matter of time before the "design conscious" crowd, which includes a large contingent of Apple employees, alights on the year 2005 as the pinnacle of "retro" and starts creating throwback versions of everything. By my estimation we're at about 1994 right now so ...
I literally have a playlist called 'modern 80s' which is 2020s music releases that kind of sound like they're from the '80s. I definitely subscribe to the 30 year nostalgia theory!
There were a lot of bands harkening back to an 80’s sound in the 2000’s, and people were saying that the “retro cycle” was therefore 20 years (if it was recurring in the 2020’s that’d be 40 years). But I think someone could make the case that the 80’s have just become the permanent retro decade, and that everything since 2000 has been subsumed into a sort of “infinite present”.
I think the internet and free universal frictionless undifferentiated unbiased access to recordings has made all sounds equally available at all times, starting somewhere in the 90's, and that has made anyone who was born after that point perceive them a bit differently than before.
I have a neice who was born in 96. When she was 10 or so I found out she was listening to both Green Day and Patsy Kline with essentially equal interest.
That's when I got this idea.
She was born entirely after not only the existense of recordings that go back at least a few generations, not only after the existense of the internet, but after the mass adoption of the internet, digital copies of recordings, and countless distribution means, both centralized and peer to peer. The essential nature was not that different from today, the day she was born, let alone 10 years later.
For me, Patsy Kline was only on the oldies station at my grandmas house and the barber shop and which actually called itself the oldies station, and my parents wouldn't be caught dead listening to Green Day.
I can go from Yes, to Twenty One Pilots, to Kansas, to TheFatRat, to Johnny Cash, to random youtuber, to Linkin Park, to Guns 'n Roses, to Elvis, and so on.
They're all just there, there's good stuff from every time period - why would you limit yourself?
I think there's a technological aspect too where after 2000 or so, nobody has really figured out as many new novel sounds to make with synthesizers or turntables or instruments or whatever. The technology of music instrument/tool development plateaued, so the novelty started to as well.
I’d definitely disagree on that front. Though I do think there’s been a bit of a logarithmic curve, the difference in what was possible in 2000 vs 2022 is pretty huge. Like, listen to Animal Collective’s Merriweather Post Pavilion (2009) or any of James Blake, Jamie XX or Burial’s stuff from the past decade. Not all of that was impossible before 2000, but you’d have a hard time finding people making music with those timbres (outside maybe some really cutting edge IDM).
Oh and don’t forget auto-tune ;) Although there are a few examples of it being used for pitch correction before 2000 (+ that Cher song), it wasn’t until the 2000’s that it got ratcheted up to T-Pain levels.
How does the magnitude of that difference compare to vs 1980, and then to 1960, though? Both in terms of the possibilities and in terms of how much the possibility space had been explored?
>But I think someone could make the case that the 80’s have just become the permanent retro decade, and that everything since 2000 has been subsumed into a sort of “infinite present”.
That may be true culturally in general, but my theory is just that the 80s marked the end of the period where the sound of pop music was heavily constrained by the technology of the instruments. After that, synthesizers got good and didn't have to sound like synthesizers. There was a little gasp of that when Autotune got overused, but otherwise, music can sound however you want it to sound now, so there's no consistency of style.
It used to be 20. 70s—>50s (American Graffiti, Grease, Happy Days), 80s—>60s (Summer of Love, Touch of Gray, Love Shack), 90s—>70s (Tarantino, lowrise bell bottoms), 00–>80s… When you’re 18-25, your model is the previous generation. Around 30, the pattern resolves and your grandparents somehow become sophisticated in hindsight.
I miss a lot of things about the 10.4-10.6 era. Brushed metal Safari is not one of those things.
As I recall, per the HIG, the official line on brushed metal was that it was for apps that interacted with physical hardware in some direct way: portable music players, synch. managers, optical drives, disks, something. Now if you think that’s vague, you’re not wrong; most apps could at least print something, and it wasn’t applied consistently even within the iLife suite which up until the inclusion of iWeb consisted entirely of apps that were intended to bring the Mac and peripheral devices together (CDs, digital cameras, MP3 Players/iPods, DV cameras, SuperDrives, MIDI). This vagueness and the low popularity of brushed metal Safari in particular was one of the reasons the brushed metal theming was retired after Tiger.
I know I'm in a minority with brushed metal but I really liked it, yes its official usage guide was vague and it probably was abused a bit but from a purely aesthetic point of view I liked it, like a lot of things in Aqua of that era it made things feel very tangible. I'm one of the apparently few people on HN who actually liked the touch bar (but not the lack of a physical escape key) on the MacBook Pro as well so take my opinion with a pinch of salt!
That’s fair, but I found it grating in any app I looked at for a decent length of time, like web browsers. I think brushed metal with stricter usage guidelines (followed by Apple) that wasn’t per App but per Window-type could have stuck around. Having some way to distinguish Apps and window types in Exposé isn’t that worst thing in the world and was about the only redeeming aspect of Apple’s Lion theming choices.
Anyway, I used Firefox and Camino around that time.
Only on a MacBook Pro or PowerBook. Not accounting for 3rd party accessories: iBooks, MacBooks, Apple’s Mice & Pro Mice, Mighty Mice, Apple’s Pro Keyboards, Apple Keyboards and Apple Remotes were all plastic at the time, and mostly white or translucent plastic with a couple of exceptions.
The aluminum stuff all came about just immediately prior to (about one or two month’s prior) or after Leopard’s release, but Tiger’s era (and Brushed Metal didn’t start on Tiger) was dominated by plastic input devices.
Which is pretty incredible, since Tiger was the first public version of OSX to have distros for both PPC and Intel!
Tiger was solid - Leopard was where it was at. Snow Leopard may be my favourite, but as I can’t run it on any of my PPC systems I have no excuse to run it anymore.
I actually liked Exposé a lot more in Tiger and Leopard than in Snow Leopard, where all windows were resized to a grid, and you lost relative window sizes as a visual cue. Thankfully, there was a hack to get Leopard Exposé in Snow Leopard, by overwriting a binary with one from a beta! https://superuser.com/a/212717
There were hacks for absolutely everything. Simpler times...
Me too.. The old expose was way better. I also loved the virtual desktops in a grid.
When they moved them into a row and changed the way multi-screen worked it completely lost its usefulness for me. Before that I always used a 3x3 grid. It was one of the many ways macOS evolved against the grain for me.
Other major regressions: LaunchPad (cool for iPad users I guess but completely ridiculous on a 24" screen - Even Microsoft moved back from this). The incessant power saving tricks imposed that are all but impossible to switch off (I use mainly desktops so I care about performance, not battery). Notarisation, every app calling home to Apple every time it's run.. Wtf. Then there was Apple "Server" which became more of a joke with every release (the Leopard version was the last real one).
But really what it told me was that I hate an OS where every year some people sit around a table and decide what I should want changed, and offer no way to choose. In other words: I hate opinionated software.
I recently moved to KDE and wow, it's so amazing to have all the things I missed back. And also a more useful expose-style function again though the relative window sizing is unfortunately something it doesn't do.
I actually still use a quad core G5 with 16GB(!) of RAM, and 2x1TB SSDs for some audio production work in Logic 9.
64-bit started for me back when the G5 came out, and today that G5 screams, easily feeling as responsive as my girlfriend’s M1 MacBook Air most of the time.
It’s almost unbelievable how the G5 performs like it’s got bloody Sonic the Hedgehog trapped in there on a hamster wheel generator. My Intel machines never came close, and a lot of it seemed to have to do with major bloat as the OS moved along. Probably also bloat in Logic. My G5 can handle the hell of a lot more effects and VSTs with about 50% the impact they have on my Intel machines. It’s surreal.
I also find it absolutely pathetic and astounding that up until this year (16 inch MBP excluded) the maximum RAM a MacBook could have was the equivalent of a computer I got in 2005. 15 years later I still couldn’t get even 32GB RAM in a MacBook. And it was soldered so it couldn’t be upgraded. Shameful.
The 8 DIMM slots don’t fit in a laptop anyway ;) Seriously though, the 16GB limit was mostly due to the Intel parts not supporting more LP-DDR.
But yeah these things were beasts. The dual-CPU and then the dual dual-core were seriously impressive. I really wanted one at the time but could not justify it. I finally got one for €100 2 years ago, now it sits next to a G4 Cube. Both are some exceptional pieces of engineering.
It is real Snow Leopard with a Snow Leopard kernel, but a lot of components are copied over from Leopard, and there are still a lot of issues. I'm not clear as to whether or not anyone has gotten graphics acceleration working, for example—it certainly doesn't work on most models.
I was one of the Aqua haters, but I got used to it quick after installing OS X Public Beta and switching it to the graphite color scheme. Doesn’t mean I don’t miss OS 9 Platinum, though. I’d be interested to hear from anybody who spent time using the Dark Platinum/NeXT hybrid desktop of Rhapsody and OS X Server 1.0/1.2, although I have to wonder if there was ever anyone who used the server OS as a daily driver.
Fun fact: OS X Public Beta had an Easter egg setting that I can’t quite remember, defaults write com.Apple.something or other, that would set all Cocoa apps to a straight-up NeXTSTEP appearance. It completely broke usability, though, as windows would minimize into little squares in the corner of the screen a la NS rather than into the dock.
> Fun fact: OS X Public Beta had an Easter egg setting that I can’t quite remember, defaults write com.Apple.something or other, that would set all Cocoa apps to a straight-up NeXTSTEP appearance. It completely broke usability, though, as windows would minimize into little squares in the corner of the screen a la NS rather than into the dock.
Wow I've never read about this -- know of any screenshots? Would be really interesting to see what some of the default OS X apps looked with the NeXTSTEP appearance.
It’s not exactly what you’re asking for, but this Twitter thread from Stephen Troughton-Smith has screenshots of several stock OS X apps as they changed from NeXT/OpenSTEP → Rhapsody/OS X Server → Developer Previews → Public Beta → final release.
You could move/rename the `Extras.rsrc` file in `/System/Library/Frameworks/Carbon.framework/Versions/A/Frameworks/HIToolbox.framework/Versions/A/Resources/` and disable Aqua in the very early versions, returning lots of things to a Mac OS X DP2-style Platinum appearance:
I loved Aqua, particularly after they toned down the pin stripes and transparency, so, yes, around Tiger (and up until around Snow Leopard). It was indeed a great step forward in terms of UI, which cannot be said of many more recent releases.
Everybody is very positive about Tiger’s UI, but what I remember were the disjointed elements. Brushed metal was prevalent in many apps, and completely broke the UI’s look.
Even though we have many gaping holes in the details of the current UI due to the many paradigms (SwiftUI, AppKit, Catalyst), skin-deep macOS is much more agreeable.
Yeah brushed metal from the experiments in Panther was still too present. It got better with Leopard.
Cocoa apps are mostly fine in modern macOS. Their main issues are the lack of contrast*, but things like the dark theme are great. From what I have seen SwiftUI is fine too, although it still has some way to go to catch up with AppKit, and has some rough edges. Catalyst is plain garbage, though.
* a pet peeve: I have to click 3 times when I want to make sure that the shuffle mode is either on or off in Music. Another one: document proxy icons are getting hidden and harder to use, even though they are a fantastic feature of the OS. There's a bunch of others features that are becoming more and more obscure, which is a damn shame.
Project manager driven decision-making. Everyone wants to make a name for themselves, so every update gets a big UI overhaul even when it doesn't need one. Features get taken away when they don't need to be. Workflows get redone when they don't have to be redone. Imagine if this were to happen for something as important as Bash, people would revolt and fork, but in the big tech world you are beholden to these corporate products.
It's not just apple, its everything in big tech. Venmo just released an update where they moved key workflows around for no reason at all, and now I don't have muscle memory for the app anymore, but I'm sure some project manager justified it with telemetry and got a huge bonus for rolling out an update and showing downloads grew by 1% (which they probably would have anyway).
It's rose-tinted glasses. Every release meets the same same level of criticism as to how it's the end of macOS, how the interface design is a regression towards infantilism. It has been ever thus, even in the System/OS 7, 8, 9 days. The difference now is that macOS is more mainstream.
Mh, I don't know. Some of the changes that people complain the most about (from Leopard's 3D dock to Safari 15 tabs) were undone by Apple. And for some changes that haven't been undone, like monochrome icons everywhere, I still mis-click things so often that I doubt it's just nostalgia.
If you'd let people mix and match elements from different eras of macOS, I'm sure you'd see some patterns regardless of when people got into Macs; similar to how people have lots of abstract opinions about architecture, but somehow the tourist buses always stop at the same cozy-looking old towns. Beauty and usability are not entirely subjective.
I’ve been an Apple user since the early 80’s. Apple being simultaneously the gods of UI and scourge of UI have been a constant, along with the ‘doomed’ narrative. A lot of this is the peanut gallery repeating what they’ve heard, as well as exaggeration of the issues individuals face. For instance, in the 40 or so years that I’ve used computers, I believe they have never been more user-friendly than they are now for the typical user. Not just Macs, but Windows and Linux’s desktops too. Reading opinions here would make you think the opposite is true, but here is full of people that love to tinker, and fewer seem to want to go back to the “good old days”. Hence the ‘rose tinted glasses’ comment. I was there, it wasn’t that great! I jest, well a little bit anyway. I remember Tiger being released and a-not-insignificant-amount of people complaining about brushed metal. As I said, it has been ever this. Long may it continue - it makes us that do care think.
Yep. Everyone is cool with all the change that was necessary to get software to the point where they were most emotionally attached to it, and all change after that is “useless meddling that no one likes and is only done to give PMs and designers a job.”
This massively oversimplifies the state of technology and is unfairly dismissive of the significant productivity hit that some changes can inflict on the most loyal and experienced users.
It’s certainly possible for changes in UI to significantly harm productivity, but you would need to systematically gather evidence to know if this is happening. Whether it’s happening is almost completely independent of anyone’s individual feeling of frustration at needing to learn about and adapt to new UI changes.
But more importantly, they existence of change is inherently important to the large-scale advancement of computing over time. Even if it is the case that a certain UI overhaul of a major operating system harmed productivity, the solution is not “permanently halt all software changes after this specific version that I have learned and enjoy using.” We ought to reject arguments that forced stasis is the solution any time changes introduce risk.
Look at the MacOS release feature lists before and after the iPhone. It's pretty clear that the Mac teams got cannibalized and didn't recover for years.
It’s not universal, people just don’t bother writing posts with the opposite perspective because it’s boring. Tiger was my first version of OS X and I think what we have now looks much better.
The UIs are designed for the monitors of their time and it’s not fair to compare them on the same screen. 2000s computers had shitty TN LCDs with low contrast which is why the UI had so much kitsch.
They are in the Windows 8 phase of design. They want to merge mouse and keyboard systems with touch systems, and are thusly forcing mouse and keyboard users to use interface conventions that are derived from touch platforms.
Mobile platforms benefit more from high contrast, very simple and flat designs.
Beyond just the touch screen convenience features -- A significant portion of users is seeig their UI under conditions of extreme sunlight, water droplets, or cracked screen at any given time. These things all inform design choices for mobile.
Now then, why is Apple making the same mistake as Microsoft Windows 8 by forcing these design elements onto Desktop and Laptop market segments? Apple doesn't really think of the PC market much. According to their financial reports, they make more money selling chargers for their mobile devices than they do on the entire PC market.
This. macOS has been in an incremental phase for several years now. Windows 8 was a major overhaul from Microsoft which ended up as a big mistake since they didn't consider the impact of such an overhaul in terms of user experience. macOS on the other hand is quite mature and Apple knows that any major change would hurt users in the end.
Just an "incremental" update where they added touch screen apps (from the iPhone and iPad) to the desktop and laptop experience. Oh, they also changed the system tray to mimic the touch functionality of iPads 1:1 as well, despite being a (far more precise) mouse pointer controlled system. Incremental they say. I was way off base and taking crazy pills when I saw any relationship between these design elements and touch interfaces.
To be honest I really kept away from that old Mac interface for years. When they switched to a cleaner and flatter design, I literally went "now I want to jump on Apple ecosystem". Same for iOS 7 update.
They weren't the only or primary reasons but they had significant role. I personally love the new flat interfaces MUCH more TBH.
I couldn't really stand the tiger-era MacOS either; but early iOS and *MacOS Snow Leopard* we're the pinnacle of UI/UX for me.
I still remember the high pixel density of the devices being shown off so well with the crisp rendition of paper and leather. The way it felt like it was popping off the screen.
Back then I had really good eyesight (I was 20-22~) and those UI elements sold me on the quality of the hardware.
That, and it was much smoother in it's animations than android/windows/compiz.
All of them were the right choice for their time though. Especially early iPhone and its skeuomorphic design taught (practically) the whole world how to use a touch interface with buttons, lists etc. and it looked great for the time. Similar for old Macs too. Then flat design, IMO, cleaned up the general UI after teaching it.
Absolutely. It feels like playing battlefield. Just like you do not see the soldiers in that game, you don't know where the buttons or clickable things are. Does this page when scroll? Who knows. Maybe it does.
I still have a working 2007 16” MacBook Pro (fully loaded 4GB of RAM, yeah!). It’s not just the software that was better but the hardware too. That keyboard has so much travel compared to these newer ones, it’s crazy. It actually feels like a real keyboard.
The newest generation of Apple keyboards has a bit more travel than the previous generation, but the laptops from 15 years ago are nicer. None of them come close to the glory of this keyboard though: https://en.wikipedia.org/wiki/Macintosh_Portable
It is. Also really heavy because it used a lead acid battery. It was also wired in such a way that if the battery died it wouldn't turn on anymore unlike modern laptops that can run off the AC adapter directly. My parents threw theirs out because of that.
I still have one of those too. I do love the keyboard on that, it feels very luxurious to me, but I have to admit I can't type as fast as I can on my 2019 MBP work or 2021 M1 personal MacBook Pros.
I'm definitely in the minority in that I actually prefer the 2019 keyboard over the others. It's the "fixed" version with the rubber gasket that prevents dirt getting in the works and I've never had an issue with a stuck key. Maybe it's because I'm a very light typist, but the 2019 butterfly keyboard never leaves my fingers feeling tired after a day of typing, where the 2007 keyboard did.
15" or 17"? I have a PowerBook G4 from the same era, and I think the keyboard is the same as that. Indeed it has better travel, but I find the keys are slightly wobbly and a bit unrefined. Having torn one apart a while ago, I'd suspect it's due to looser tolerances than more modern designs. I think ThinkPads from the same era have significantly better keyboards. They're less mushy and have a better texture.
The wobble in keyswitches is often intentional: it prevents keys from binding when pressed off-axis, and makes them a bit more forgiving when they get dust inside.
I didn’t like the older thinkpad keyboards; noticeably too stiff for my taste. But this IBM "portable" keyboard is a dream: https://en.wikipedia.org/wiki/IBM_5100
I just don't understand how many people prefer these keyboards.
Key travel is only important to assist in your typing + provide feedback, but unless you type like a brute (I've a friend whom I would NEVER lend my laptop too, as I would be worried for my keyboard) or have sensory nerve issues like from diabetes, you don't need much travel.
I was dubious of the newest Lenovo keyboards, but after using my X1 Nano for a few hours, I was convinced: while the keys themselves have less travel, it has a strong opposing force that gives a lot of feedback.
It's still a bit too stiff for my taste (it seems to have been made for big burly guys) and I'd prefer something more like the current Macbook keyboards, a self-avowed heresy for any Thinkpad fan :)
It's funny to think about Lenovo making the X1 Nano for big burly guys.
It's too bad that laptop switches will likely never see the huge and varied aftermarket that we have with mechanical keyboards. There's a switch for everyone, and it'll work in almost any keyboard if you're handy with a soldering iron or have one with hotswappable switch sockets.
> It's too bad that laptop switches will likely never see the huge and varied aftermarket that we have with mechanical keyboards. There's a switch for everyone
Is there one for me, who likes little key travel, softness, and silence?
So far the best solution I've found : thinkpad USB keyboards everywhere :)
Possibly a lightweight Kailh Choc linear switch? They’re very soft and quiet, but they have no tactility. The tactile ones leave a bit to be desired. It’s hard to match rubber domes there.
Indeed. Even if it's antithetical to the mechanical keyboard idea, I wish MX style keycaps (wide availability, for ex this is how I could get a Cyrillic keyboard) could be made compatible with rubber domes (with no mechanical switch)
It's a tough engineering problem, that's for sure. I agree that older ThinkPads are a little bit stiff.
Apple went too far with butterfly, but I think they landed a good place with their current line up. They have very little wobble. Cherry MX switches (and clones) are excellent in terms of off-axis binding. Some clones are significantly less wobbly than others, for various reasons but none of them sacrifice off-axis performance. It took decades to get to this point though.
On a related, Alps mechanical switches are notorious for dust and dirt ingress issues as they get older and they're extremely hard to clean. For some reason Cherry MX switches have fared much better over the years.
> Alps mechanical switches are notorious for dust and dirt ingress
This is not accurate. You are thinking of cheap Alps clones. The “complicated” SKCM/SKCL switches from the mid 80s through early 90s have fancy switchplates that give an extremely clean electrical signal and have contacts entirely enclosed in a switchplate. All sorts of 20+ year old keyswitches (including Cherry MX switches) start to feel “scratchy” when they get used with a lot of dust/dirt inside, because the plastic gets scraped up.
I am not thinking of clones, no. I’m thinking of my AEK and AEK IIs with creams and salmons that don’t feel so hot today. I haven’t used many vintage Cherry boards, but I do know that people actively seek out “vintage blacks” because the plastics have been smoothed out after a lot of use.
I’m really surprised to read this comment because it really disagrees with what I’ve read and my own experience.
> “vintage blacks” because the plastics have been smoothed out after a lot of use.
No this is not why people want the older Cherry MX switches. The older MX switches used less worn-out molds and/or different plastic, and were much smoother straight out of the factory than MX switches from a few years ago which are quite scratchy feeling when new (this may have improved more recently, I’m not sure).
> my AEK and AEK IIs with creams and salmons that don’t feel so hot today
Feeling scratchy is very common for 25-year-old MX keyboards as well, and other kinds of keyboards. Stuff that sits on a shelf or table (not in a box) for decades collects a lot of dirt inside. Alps also applied some kind of lubricant to the slider in the factory, which may not be there anymore. Alps switches might be a bit more susceptible to scratchiness from decades of dust inside than MX switches because the slider slides against 1 or 2 leaf springs which provide resistance; in linear MX switches the slider is only moving up and down in a plastic channel and scraping a bit against the sides but the resistance is all provided by a helical spring.
(You could probably improve your key feel by disassembling all of the switches, putting the parts in an ultrasonic cleaner, and applying some lubricant when re-assembling them. You might not consider that to be worth the effort though.)
But I would generally expect both of your keyboards to be electrically quite reliable (at least at the switch level, with a very clean signal across each switch, not much chatter).
If a keyboard is stored for an extended time with key(s) depressed, it can end up deforming the springs. Deformation of the leaf springs can reduce the switch resistance and tactile snap in Alps switches; you can try to restore this by bending the leaf springs outward yourself, but it is pretty tricky to make them consistent.
I also have an AEK keyboard which stayed in a box somewhere sealed for 25 years or whatever, and is amazing today.
I think it’s a mixture of molds and being broken in. Apparently these artificially broken in[0] switches with Cherry’s new Hyperglide molds are quite good.
I did try ultrasonic cleaning some of my Alps switches, but it’s hard to replicate the factory lube. The lubes we use for MX switches aren’t ideal.
I would fully expect a new old stock AEK to be amazing. All of my AEKs are indeed electrically reliable, but they are scratchy.
Unused new-old-stock MX switches from the mid 1980s have noticeably smoother feel and don’t make the scratchy sound you get with new MX switches from, say, 2010.
Artificially wearing in keyswitches with 100s of thousands of robot keypresses seems really silly to me, though I guess it’s easier than disassembling every switch and sanding down the friction points with fine-grid sandpaper or applying fancy aerospace lubricant or whatever the kids are doing these days. YMMV.
Retooled switches from 2010, yes, but Hyperglides are new moulds as of 2020. Cherry made a fairly big deal out of it at the time. I've heard of people using diamond polishing paste to do something similar.
I've got a 2009 MBP, After continuous use enough of the keyboard domes finally cracked to become (almost) unusable after 10 years (made it Linux after 5 after Apple abandonment), now a lot of the keys are mush but still actually work.
i duno... "they don't make em like they used to"? :P
Apple's post-Steve brain drain is real, and a lot of the internal redesign projects are now simply resume-driven development.
A lot of the people working on these systems weren't working professionally when 10.0 came out. Most of those people have moved on. Apple is not the same humans.
A corporation is a shield behind which humans plot and practice their most deliberate, libidinous schemes of avarice, moderated not by morality or community, but only by laws and markets. And for that, corporations have lawyers, resumes and advertising.
It is a rare corporation that retains a soul for longer than its founders presence, because lawyers, resumes and advertising cannot sense what founders could: a creative future.
I would think an increasing amount don’t remember it either. I was three at the time. The earliest OS X version I remember hearing about was Tiger, and I used Leopard.
Definitely. It feels lively and exciting. Jonny Ive's decision to go with flat killed all joy in the UI space and millions of designers copied it mindlessly which I find especially egregious.
Wow, Agreed. The window shadows here really help them pop way more than I remember, and they seem to do a much better job delineating overlapping windows than the current “styles”.
macOS still has window shadows and they are still the same size (by eyeballing them). The only difference is that in dark mode the shadows are of course less visible because the contents of windows are darker.
At some point, window shadows became much more diffused. I think the release that implemented this change was mountain lion.
I keep a Snow Leopard VM around to occasionally run old software in and the difference in shadow size is always striking. I personally like the more focused look of pre-10.7 shadows.
I just took a look, and I really can't see any difference between the window shadows on my Mavericks machine and my Snow Leopard VM. The Mavericks shadows may be ever-so-slightly stronger.
You may be thinking of Yosemite, which definitely toned down the shadows a bit as part of flattening everything.
I think, in general, UIs back then felt more "human": soft, intuitive, yes, sometimes a bit clunky, and yes, sometimes with contradictions (like any human being). Now they feel more "robotic": harsh, darker (or is it that designers nowadays tend to use softened colors?), precise as any machine (but this doesn't always mean that they are easier for humans to use). Robotic, though, doesn't seem too far off: society is becoming more and more robotic I believe.
(In case there needs to be a reminder) Take a look at hellosystem if you want this kind of interface on a modern OS(FreeBSD, that is).Yes, it's nowhere near a complete solution for most work or even some average usage, but it looks promising and we don't deal with apple.
I also found the older UIs of Tiger and Snow Leopard to be more intuitive. Modern flat UIs are difficult to discern what is presentation and what is intractable. Buttons used to look like buttons. It was obvious back then that you can click on them. Today, buttons are so flat that you have to think harder about whether what you are looking at is a button or a text label with a backdrop shape.
Any method I can think of to replace UI assets would be prone to breaking with every patch Apple pushes out. I’m not going to say it’s impossible, but it’s probably not worth your time. If you want to explore what the UI felt like, you’re better off exploring VMs or old hardware and install media.
To be honest, with the way Apple does support for releases 1 or 2 back from current, you could potentially do this easier nowadays: just don't support the current version (e.g Monterey), only target Catalina or Big Sur. When Monterey is no longer the new one (only receiving security patches), that's when you roll it forward.
Scott Forstall was fired for the Apple Maps mess up. He championed good designs like this. When he left it was a race to the bottom at Apple for their UI/UX.
Scott created a lot of friction and animosity amongst various high-level managers and executives. Steve protected Scott. With Steve gone, the other executives had to deal with this issue. I won't name who, as I am sure the story has been told elsewhere, basically said Scott had to go or they would go. Scott may have gotten Apple Maps pinned to him on the way out the door, but there were many, many more issues leading to his departure.
I always wondered how early Mac OS X developers learned to navigate XCode and Cocoa. Were they all just ex-NeXT programmers?
I bought an old copy of the Hillegass book Cocoa Programming for Mac OS X, but I'm guessing most people learned from Apple developer docs that I never saw (or guess didn't know how to find back then)
As an aside, I don't miss the old programming books where each chapter just showed you how to use some GUI elements and they never got around to showing patterns on how you would actually create a usable application.
Ha, in the 2006-2010 timeframe I worked for an aerospace company working on experimental radar sensors and the like and I developed quite a few Cocoa desktop data analysis and visualization tools since we were a Mac shop.
It was a good fit since the physicists and mathematicians wrote their experimental stuff in Matlab and the engineers and us computer scientists wrote the actual production code in C compiled for the embedded hardware in the avionics.
Cocoa provided a good bridging platform for tool sets between the two camps. (Later, python and its robust set of science and math libraries became the tool of choice in this role as Python expertise became more general on our team)
Long story short, I learned by suffering through the interface, especially the fairly esoteric Interface Builder.
And then in 2008 I went to a bootcamp hosted by the former-NeXT guy that had founded Big Nerd Ranch, Aaron Hillegas. He made it just click, so yeah, maybe but did take a NeXT programmer to wrap your ahead around it, ha.
This was key. I built a couple of minor, but useful for me, apps by simply following the docs at the time.
The major difference between Apple docs, and Javadocs at the time (Java was the actual language I used to program in), was that Javadocs was basically API references, whereas Apple docs had API references but they also had guides that went into the why, and the best practices, and even occasionally alternatives for edge cases, etc.
And the fact that it was available offline as an optional download with XCode was a massive bonus at a time when ubiquitous Wifi and internet was not a thing.
It was a real surprise to me, after having stepped away from any sort of Apple development for a few years that Apple's docs were considered bad.
Shamefully, as an Apple employee working on Finder/Spotlight/Time Machine and more, I never, ever used XCode. Almost no one on our team did. We all used a combination of the terminal, our text editor of choice and command line tools to wrangle together binaries for local development. The actual production build was done by the internal build system which also used various scripts. Why didn't we use XCode? Take the same reasons stated today and apply them to 2001, 2002, 2003, etc. etc. etc.
I learned Cocoa in 2002 on Mac OS X 10.1 and Project Builder (the NeXT IDE that was Xcode’s predecessor).
I remember reading a brand new O’Reilly book called “Building Cocoa Applications”. It was written by two ex-NeXT devs and was quite helpful in understanding the system.
Apple’s own documentation and sample code was good back then. The API was much smaller, so I read the reference for every Cocoa class.
Apple provided a fully functional demo app called Sketch which showed important architectural patterns like undo/redo. When developing my first Cocoa app, I’d basically go see how Sketch.app implemented something and copied the approach.
(Incidentally, I’m convinced that the well-known Sketch drawing app is basically a case of somebody looking at Apple’s Sketch.app sample code and thinking “couldn’t we just sell this.”)
Some years earlier I had tried to learn Win32. It’s hard to overstate just how fun and easy and powerful Cocoa felt in comparison.
> Apple provided a fully functional demo app called Sketch which showed important architectural patterns like undo/redo. When developing my first Cocoa app, I’d basically go see how Sketch.app implemented something and copied the approach.
As someone who was a teenager who couldn’t afford books up through 2009 or so, most of my learning of Obj-C and Cocoa came through scraping the internet for blogposts on whatever subtopic I needed to know about. Occasionally you’d run into a full fledged tutorial, which were gold mines. At one point, I made AIM friends with a couple of people who were more knowledgeable than myself which was a great help, and later on Stack Overflow appeared which let me both ask questions and peruse the answers to others’ questions.
It was kinda rough, and I didn’t get to the point to where I could build useful things until the late 2000s and early 2010s… just in time to dovetail into iOS development (which I’ve now been doing as my job for the better part of a decade).
The cornucopia of free resources that are available to new learners today is a ridiculously stark contrast to how it was back then.
I did learn from the Hillegass book -- the first edition released in I believe 2002. I also had a couple of old NeXT programming manuals, and had done a tiny bit of NeXT work in the past. And I agree with the others who said Apple's dev docs were better back then. But really Aaron's book is what got me started.
The most useful non-source code resources at the time were about GCC, BSD and Apple’s PDF guides on how the OS works, and the Human Interface Guides, paid Developer Connection accounts got you tools mailed to you and DVDs of WWDC seminars.
But it’s a profession, selling shrinkwapped desktop applications or the modern equivalent takes a lot.
I built a decently widely used mmo chat client, Cocoa was very well documented in books. Zero NeXT knowledge from my side. IMHO QT still lacks the amount of information that was available at the time for Cocoa.
Original MacOS X ran unmodified apps for original Mac (which they called Classic) using emulation, slightly modified apps for the original Mac (recomplied to Carbon APIs) or slightly-modified NeXTStep apps (using Cocoa APIs AND InterfaceBuilder which became part of Xcode), or even slightly-modified Unix apps (using POSIX APIs) along with AppleScript for automation. There were few NeXTStep/Cocoa apps, except for simple graphical wrappers around POSIX apps. Then Apple started killing off APIs (including Carbon), changing functionality and picking winners/losers to drive the developers to retrain, rewrite and maintain their code base for Cocoa, which most did not do, leaving Mac OS native apps to languish for many years (sort of similar situation to AppleWatchOS), especially after the codebase was forked to become iOS. However, HTML5 apps were on the rise and WebKit sort of kept the Mac hardware sales going. Then, a year ago, Apple introduced the ability to run iPad apps on new Macs and rolled out SwiftUI (which, when it works, can target either iPad or macOS natively), which is sort of the final nail in the coffin for Cocoa.
yes, and.. "Apple started killing off APIs (including Carbon)"
there was some transition time when Apple published Carbon interfaces to Mac OS 9 devs (like me), stating that they were "transitional". Quite skeptical, I used them to rebuild some tools and apps in CodeWarrior. Within a short time, more updates had less Carbon, and the news came out that Codewarrior was locked out of OSX -- no deal. It was obvious that the Mac OS 9 interfaces were for chumps, and who wants to be a chump. It was true, and things changed.
yep, dangerous business calling out Apple these days tho. The only thing particularly relevant to the New Peoples on here is the takeaway that Apple has, can and will change its APIs as a means of exercising market and labor control any. time. they. want. to. It's in their DNA because their leadership came up in a time when these mechanisms were off-radar of labor regulators and the FTC (that they perceived to then be as relevant as modern day Bitcoiners surely perceive the SEC to be). The worst example was probably OpenDoc, but there are plenty of others https://en.wikipedia.org/wiki/Taligent Can still hear batshit criminals (now with personal fortunes worth $134B) talking about their "right" to innovate...(ie, F with the APIs) In looking at what's happening with Section 230 Reform, one can't help but wonder if the current dystopia could have been avoided had Congress given such attention to operating system APIs in the pre-AppStore/pre-HTML5 era.
Also, the GUI for building GUIs was great. Nothing really made you appreciate why they wanted objects so much as dragging buttons around and instantiating the class. It made sense internally to me as a university student at the time. Objective C was a "relatively simple" set of extensions (which I never really understood) over C (which I claimed to understand at the time) and the language made you aware of both "the magic" of what you were doing and, at the same time, how it related to the bare metal. I learnt a lot from it.
I'd love to know what this looks like to a straight-out-of-university developer of an Electron app, though.
Making GUIs was easy as long as you stayed on the happy path, but the moment you needed to do something different you were back to writing reams of code, there was a cottage industry of custom NSSplitView classes. Even just making a button a different color involved reinventing the wheel a substantial amount.
The current Electron/React approach has no happy path - everything uniformly requires some amount of boilerplate code. But when you need to deviate you are less likely to have to write a novella.
> Making GUIs was easy as long as you stayed on the happy path, but the moment you needed to do something different you were back to writing reams of code
I'm making this up, but perhaps this was beneficial for UI consistency across apps anyway?
Yes it was. But such consistency is double edged sword because it also leads to “if all you have is a hammer” UIs where UI elements are abused for roles that they shouldn’t be used for, because the alternative custom UI costs so much development time.
In general I’d say UIs are much better these days functionally, at the cost of much less consistent look and feel (UX) across apps.
Having implemented multiple programs in GUIs for GUIs (with GtK, UIKit and dabbling with AppKit)... I must say that I have my past self every time I want to go back and look how something works. At least with SwiftUI / reactive things the code can be searched and navigated and doesn't take ages to load.
Ahh NextStep. For some reason I was only thinking of Mac OS. Possibly because in 1988 i was a kid with an 8 bit spectrum clone who only saw NeXT machines in magazines ;)
It's not just Apple. Microsoft also stopped producing first-party documentation on low-level technology, just because they wanted to stop people from using certain things even though they were still very much a fundamental part of Windows.
The decline of 'FOSS-friendly' Apple coincides with the re-licensing of many projects under the GPLv3, which is decidedly enterprise-hostile by design. I concur that the developer documentation need improvement.
Well, GPLv3's installation instructions requirement more or less is targeted squarely at consumer electronics companies and that includes Apple. If Apple were to include GPLv3 software in iOS, the only way to comply with the license would be to significantly alter their security model to include an owner override. While there are ways that Apple could still use GPLv3 software, not change their security model, and remain in compliance[0]; I imagine they decided it would be easier to just ban new GPL software in their OS entirely rather than deal with the compliance headaches.
Apple in particular never shipped any GPLv3 software in their OS and stopped updating even v2 software. They used to be very heavy GCC users, but wrote their own permissively-licensed compiler that outdoes it in almost every way. In their defense, they actually wanted LLVM to be an upstream FSF project; but RMS famously lost the e-mail because he daily-drives barely functional ancient laptops. In a sense, that too is enterprise-hostility; albeit not owing to choice of license. I imagine that if the FSF had agreed to refactor GCC the way Apple wanted, Apple would have gone through the time and effort of GPLv3 compliance.
I'd also argue that GPLv3 didn't actually fix the TiVo loophole. It can't - not unless we're going to pull an SSPL and start writing copylefts that trip on software that merely runs alongside Linux. The way TiVo got around the GPLv2 installation instructions requirement was to make their own proprietary app enforce the kernel lockout rather than the bootloader, and prohibiting that would be very draconian.
That being said, you also should take into account the historical context of GPLv3's announcement and development. The FSF had some pretty crazy ideas, like rolling the Affero clause into GPLv3, that probably scared people into dropping their upgrade clauses even if it never actually made it into an actual FSF license document. The end document we actually got is relatively tame, but the message the FSF sent was that they were willing to ship whatever license language they felt met their personal definition of software freedom. If you didn't like any new restrictions they added to your own code, tough.
[0] Stuff that runs in a sandbox container and doesn't use private entitlements probably isn't violating GPLv3, because Apple hands out free dev accounts that let you compile and run whatever, albeit with some annoying requirements to renew the app's signature every week.
In Apple's defense, it's not a trivial ask. They have security features like activation lock that would be defeated by a badly-implemented owner override[0]. Right now, if someone steals your iPhone, it becomes e-waste because you can't reset it unless you login to iCloud. But if there's an owner override that anyone can use, then they can install a version of iOS with the activation lock patched out.
That's not to say it can't be made reasonably secure. In fact, Apple actually did it on M1 Macs[1]. The secure boot policy there includes a device-generated key that only the first admin account - the Owner account - gets, which can be used to sign new kernels for that machine only. Ergo, if you want to install Linux, you have to be logged into that Owner account. You can't steal someone's Mac, wipe it, and defeat the activation lock by installing Linux. But to get there required a lot of additional engineering work[2] - the easy path is "only our software runs on our hardware".
[0] AFAIK Android has similar security features, so they face the same threats that iOS does.
[1] T2 has a similar but less elaborate scheme. It doesn't have per-volume security guarantees - if you want to dual-boot Mac and Linux, then you have to turn off the signature check on the macOS side.
[2] Engineering work, BTW, that I'm genuinely surprised Apple put in. I imagine there were some very heated internal debates over whether or not the Mac should even have an owner override. Especially given that Apple did their darnedest to ensure that the owner override can't touch anything even remotely related to iOS. If you launch an owner-signed kernel, iOS app support turns itself off; and the Apple-signed versions of macOS actually have the same sideloading restrictions on iOS apps that iOS does. Which is particularly silly, because you can get around that by just compiling for Catalyst.
Yeah I agree it's a very nontrivial task but there is a difference between not doing it because it's hard or explicitly saying "fuck you, know your place"
>If you launch an owner-signed kernel, iOS app support turns itself off; and the Apple-signed versions of macOS actually have the same sideloading restrictions on iOS apps that iOS does.
Has there been an attempt to patch this out? if the supposed owner override is actually real then technically the only thing separating you from a non treacherous ios subsystem is a few conditional jumps right?
GPLv3 addresses definitions (such as what constitutes source-code), software license compatibility, software patents as well as tivotization. The clauses around software patents are what I was referring to. By design, they are hostile towards software patents. Hence for instance Apple not updating BASH for so long (macOS still ships with BASH 3.2.57, the last version that was GPLv2) and the switch to (the MIT licensed) Z shell. Not commenting on the legitimacy of software patents, Apple's take on using GPLv3 licensed software - merely stating cause and effect.
I believe that most corporate legal departments (who are the ones controlling what licenses can/can’t be used, not the engineers) received GPLv3 as an end to simple ways to guarantee that GPL had no possibility of bringing legal trouble.
Even if it’s technically possible to comply with GPLv3 without open sourcing proprietary code, there’s enough caveats/hoop jumps involved that it was seen as too risky to even try. They don’t want, “no legal issues as long as”, they want a flat, unconditional “no legal issues ever”.
I do too. It was around this time I left my kiddie days as OS X dev behind and became an adolescent Linux dev. :) It's actually true, been using Linux since then, it was just a much better hacker environment.
OSX 10.4 and 10.5 were marvelous operating systems for so many reasons. Particularly 10.4 in my view due to the compatability layer for legacy MacOS binaries, AND big binary feature that made the same OS usable on PPC AND Intel X86. Also Quartz Composer, which I find really interesting and awesome [0].
> You’re amazed by the brand new Spotlight and Safari RSS, you like your new OS so much you want to develop apps for it.
when i was younger i had a used next slab. it had the same effect, everything was so cool you just wanted to build things for it. next thing i knew i was coming in on weekends to build a from scratch port of my then employer's product. my boss at the time was blown away.
it's no wonder to me that berners-lee wrote the first version of worldwideweb on the next, nor carmack with quake...
edit: i guess it was doom. it was a long time ago!
Maybe I am missing some context here as I have never developed for an Apple platform, but what is the point this blog post is trying to make? That the tooling used to be very complicated?
Actually the opposite, likely. The tooling was preinstalled and ready to go, and you could largely click your way through building interactive experiences.
Compare this GUI-driven, mostly-discoverable flow with the incantations you need to know to set up a React interface, for example. It's night and day.
When OS X was still distributed on optical discs, it came with a dev tools disc for the first several releases. That how I personally stumbled upon Project Builder, Cocoa, etc. I may have never found it if I had to seek it out and download it.
What does macOS / iOS / iPadOS development look like today? Do you not have XCode with a GUI-driven workflow anymore? I only poked around iOS development during the iPhone 1/3G days, but I thought tooling has been constantly improving since then.
Disclaimer, I'm only a wannabe Apple platform developer, but the GUI-driven workflow is still the predominate way to develop for Apple platforms. The most radical change is the introduction of SwiftUI, which has you writing interfaces entirely with code, rather than using drag-n-drop components with Storyboards.
There’s quite a few iOS devs who ditched storyboards and XIBs years ago in favor of full code. I did several years ago, largely because Interface Builder became so much worse after it was merged into Xcode. That, and XIBs and storyboards suck to have to deal with merge conflicts on.
I still use XIBs when doing personal Mac Cocoa development though, because the experience there is still decent (though not as good as it was). Won’t touch storyboards with a ten foot pole though, they slow down IB too much and generally aren’t a good fit for desktop UI paradigms.
honestly if you are doing native development and using the iOS/macOS tooling directly it is not that different than this. You basically have options, xibs and nibs like in this blog now have a new friend Storyboards that encompasses multiple screens and transitions but development is still GUI based and uses outlets like in the blog. A lot of programmers just refuse to use this stuff and just do everything programmatically. new kid on the block is SwiftUI which is a much saner not XML based declarative way of defining UI components like React but is still rough around edges and will likely be for a couple years.
The main difference is that now most development is actually not based on the platform but carried out with web tech. Which, compared to desktop tools, is light-years behind in usability.
I usually praise Apple's UI for often being intuitive and elegant. But it looks like, in this specific case, there is a problem only outsiders can see clearly: ctrl+clicking an icon to an UI gadget is neither discoverable nor intuitive!
When I improved the Anjuta-Glade integration, I made some effort to do something simpler:
- Open the .ui then the corresponding (by marker comments that are automatically created when the project is created) .c and .h files... boom! They are automatically associated.
- Add an ID to a widget you want to access programmatically, double-click it on the inspector... boom! Code for accessing it as a member of the "private" struct is automatically created.
- Add an onClicked signal to a button (which already has an ID), double click it... boom! Code for the callback is automatically created.
It is a shame Anjuta+Glade never became as popular as they could.
Last time I tried, Delphi had no concept of "outlet", every widget is accessible by the form class that contains it. On the Anjuta+Glade integration, you double click the widget on the inspector to add it to "private" struct, so you can access it on the callbacks.
Yes - Delphi and C++Builder's Object Inspector still does this. Double-click a control, and the default event handler (OnClick) will be created. There's a link auto-set between the button instance's event handler (method pointer) and the method, which is auto-created in the form the button belongs to.
I was absolutely expecting a comparison with how hard it's to get started and develop software on modern MacOS and XCode today (is it? I haven't used it).
Alternate take: if you want developing for OS X [sic] to be this easy, don't even think of trying to use anything other than the tools we give you (after you ask for them).
Wow, that old Xcode UI with "Active Target" and "Active Build Configuration" made a lot more sense than the current layout! I always thought that weird "Scheme" stuff was a left-over from the olden days, but it actually seems to be an intended feature that was introduced at a later time.
Tiger was a solid release from Apple that made me switch to the then-unstable Windows ecosystem. I guess back when you paid for software, Apple made sure to squash bugs so you were happy with it. Nowadays...
About 20 years or so ago, I actually got Apple to replace an out of warranty motherboard for a paid os upgrade.
Can’t find a reference to it now [1] but there was some sort of somewhat known issue with the powerbook g3 that I had at the time that presented under os x, but not os 9.
I argued successfully that it should be fixed under the “software warranty” because it said it was compatible with that powerbook, but the processor issue made it incompatible, and it worked.
It’s kind of remarkable there weren’t more issues than that considering Lombard was their very first “New World” portable, “PowerBook1,1”! Even the OG toilet-seat iBook comes afterward as “PowerBook2,1”: https://macintoshgarden.org/apple-powermac-line-of-computers...
That New World switch is such an Earth-shaking change and it’s basically forgotten. It was Steve Jobs’ first big software transition before OS X was even decided on as the future of the company.
Tiger was good. Around this time I realized apple really wanted to decide how my music and photos were organized and weren't going to let go. Their control put me off it all. Fortunately I discovered Linux around this time.
My descent into Linux started with trying to change the background image behind the login prompt on OS X. You could do it, but it required a bit of hacking. I somehow took great offense, and... that's how I ended up a programmer.
Nowadays they think there's a need to release a major update to a feature-complete product every year because marketing said so and because project managers need something to justify their existence.
That was my first encounter with Mac and I still vividly remember the 'Aaah, so thats how a computer should operate!'. And the 'just works' feeling as well. No fiddling around with zillions of settings (which actually I wasn't against at all at the time but using Tiger I didn not miss those at all, I just worked with computer instead of working for it) and the flow of the actions just made sence, visualization actually had functional purpose, helping the awareness of what happens. First time I appreciated the drag and drop, very first time, after the clumsiness of the Windows system. First time the desktop felt natural to use in combination of hot corners. It was just a deep breath of fresh air.
Too bad the experience is diminished and what I still like were there with Tiger already in almost all cases (dock, top menu, hot corners, simplicity, ...). The simple but powerful things. New things like notification center are more intrusive than helpful/informative (first thing I try to disable in a new OS with more and more effort and fiddling around and less and less success. It is forced on me!), the AirDrop is too temperamental to be reliable, iPhone connectivity is the joke of connectivities and basically awckward for simple things even, icon redesigns are counterintuitive (still have to look for photos and system preferences 4 times more than earlier, simply cannot get used to those, are not discoverable or prominent)... are just some random items come to mind suddenly, there would be more. The forced new features sometimes does not play nice with each other. Did you try switching keyboard layouts (something that worked very well for more than a decade now) with fn button while the mouse is in the middle of the screen? Based on the location of the mouse the result will be random (floating menu item underneath the mouse will take precedence over the next layout, which is the expected way), and it is there without fix for almost years now?! A completely unnecessary notification from the notification center about dnd mode activation in locked screen (newest thing pushed upon me after latest update) has no close button but covers the pop up menus of the menu buttons available there, need to log in to disappear. Just two of the annoyances the new 'features' bring. The overall feeling is that Mac got much more stuff but did not got much better, sometimes even worse by intrusive and labour intensive disabling of things I never missed and are in the way, forced one me. Starting to work for the computer again rather than the other way around.
Concerning squashing bugs of new releases I recall Mac veterans around Tiger having the golden rule: 'Never install the new version! Wait for the first fix'. So I guess this part did not change much since.
I have occasionally tried to put together a basic Mac app. I picked up the basics of Swift easily enough, but I get stuck because XCode seems completely incomprehensible to me, and my basic attempts at finding a "cheat sheet" or "idiots guide" meet with failure.
It's at this point I get frustrated enough to give up and maybe try again in another year.
Is it just me? Does XCode really lack basic documentation?
Xcode has plenty of documentation, both first-party and third-party. However, I have yet to really find anything that really explains how you are supposed to use a particular part of it, rather than just instructions on what it can do. Xcode is one of those IDEs that is pretty smooth to use if you go with how it’s “supposed” to work and if you don’t it’s extremely rough, and there’s really no good, comprehensive guide that tells you how to do this. My best advice for you would be to watch various WWDC videos about the part you want to work on, where you’ll watch Apple engineers use it on stage. They’ll usually teach you a few things about the “logic” behind it and you’ll be able to use it productively after that.
it isn't just you, I've been doing iOS development for the last few years which generally gets preference for docs these days but I have still found over the years even if most things have documentation 1) it is pretty difficult to find 2) it is lacking. A lot of times you are better off going to 3rd party tutorials to get what you are looking for, I would recommend the Big Nerd Ranch books and Ray Wenderlich tutorials. I don't know if I have come across a general Xcode intro tutorial, maybe because Xcode is massive, most stuff is task oriented and you pick up the Xcode quirks along the way.
If you tried to build a Mac app based on Storyboards, which iirc has been the default for the past few years, then yes. All the docs are for iOS and the Mac side is completely under documented.
The classic way of building building apps is quite well documented though. It has been modernized quite a bit since Tiger and Xcode 2.0, but the general structure and workflow is the same as in this article, and there’s plenty of docs.
The new SwiftUI stuff is still under heavy development, and starting with the wwdc sessions is probably the best approach.
It’s been a while since I tried it, but iirc you had to context switch between XCode and AppCode if you wanted to use storyboards or whatever they’re called.
GUI-driven development is a perfectly fine way to getting started, which is what this guide does. However, does anyone else get an uneasy feeling about developing software that revolves too much around using a GUI/heavy IDE? You can argue that Visual Studio Code is bloated, etc., but at the end of the day, I can clone a project on GitHub and easily run a CLI command to compile and try a project, and maybe use Vim if I have to do a quick edit. Any time I see a development guide that depends on screenshots or requires a specific IDE it makes me feel uneasy, even though I know that's not how most newcomers feel.
All this guide shows is an Interface Builder, basically. Either you use one of these, or walk through a boring process of setting up controls and their properties and their bindings in code. Eventually IB just interferes or breaks your layouts, but at this time you have a good visual idea of its workings. This is a good thing, even if you're a senior newcomer. What they've got wrong is that IB is not a tool, but only a tutorial.
I don't really mind it if that IDE works seamlessly without issues, is freely available, and the project data is all in plaintext. I wasn't programming in 2005 but I have reason to think the Xcode experience really was that tight at the time
When I lose faith in the IDE and its abstractions, that's when I get really antsy for the ability to opt-out/do everything directly without its help
Tiger also was the debut of Dashboard widgets, which opened up creating app-like experiences for those who knew HTML/JS/CSS instead of Cocoa (myself included).
Also it introduced the canvas element and opened the way for HTML 5.
To this day I miss being able to enter Dashboard with a single keypress like in the old days. I used that mini calculator, note taking app and weather widget constantly.
By I seem to be alone. For dome reason most people hated Dashboard.
I’m with you here! I still miss Dashboard to this day, and regret that such a great idea never caught on. Having widgets merged with Notification Center (!) is such a worse experience, especially on smaller screens.
Never mind the fact that apps like Calculator only allow a single instance/window for no reason. Only way around that is to literally duplicate the app!
I didn’t hate it :-) I actually made the Gas widget, which fetched local gas prices. It was a popular widget at the time, and ended up on a slide (amongst other third-party widgets) during the 2006 WWDC keynote.
In the vintage apple groups and subreddits Tiger is preferred over Leopard because it runs on more hardware and it's the last version that allows Classic mode that runs classic Mac OS applications.
Tiger was the first release of OSX that I was truly proud of. I came to Apple, not as part of the NeXT acquisition, but from the post-pivot Be, which had decided to focus their attention on "internet appliances". My love at the time was operating systems and specifically GUI libraries and components.
Apple internally at that time was frightening. Coming neither from Apple or NeXT, I has an interesting position, being able to talk to various people more candidly. The Blue [1] team (System 7/8/9) on the second floor of the IL2 building seemed to be in constant distress. The ATG [2] team on the 3rd floor of IL3 was being swept out in mass layoffs and departures. There were still factions of Pink [3] and Copland [4] adherents trying to get their technology into the "Beaker" builds of what would become OSX Cheetah. The Beaker builds at the time were roughly re-skinned versions of NeXTStep and pretty uninspiring.
After my experience at Be, I really wanted to be involved in creating something great that would ship and be of real value to users. At Apple, I discovered that I just wasn't happy trying to exist in the chaos. Steve wasn't yet CEO, Avie and Bertrand were establishing a new OS organization on the 4th floor of the IL2 building and Steve Glass was still fighting to keep "OS 9" alive. In fact, OS 9 was critically important as it was needed to run on the new iMac and support all of the Apple hardware that was bringing in (diminishing) revenue. On that note, Steve was actively batting the Mac clone makers (or leeches according to Steve.)
In a moment of bleakness I received a call from a friend from Be. He said I should come join him, Andy Herzfeld, Susan Kare, Bud Tribble, Bart Decrem, Stan Christensen, Darin Adler, John Sullivan and more at Eazel. [5] Eazel wanted to create a user-friendly Linux distribution with a services model to generate revenue. The main product of Eazel was the Nautilus file manager and contribution to GNOME. After failing to raise addition capital after the initial 10 million dollars, Eazel went through a couple of layoffs. On the evening of shutting the doors, Andy gave Steve a call and told him about the Eazel team and Steve set up a large meet and greet with various Apple teams on the 4th floor of IL2. Those who were interested went to the meeting; the majority of those who weren't, ended up joining with previous comrades who had left Be to form Danger, who were now at a startup called Android.
The group who went to the meet and greet contained some significant contributors to various Apple software and hardware efforts; Darin Adler, Don Melton, Ken Kocienda, Bud Tribble, Maciej Stachowiak, Pavel Cisler, John Harper and more. Pavel helped in convincing Dominic Giampaolo [6] to come to Apple. This group of people also convinced other key contributors to come to Apple who were leery due to Apple's past history.
All that wanted to take a job were hired on the spot and we all showed up on campus got our pictures taken and started doing whatever project we thought was cool.It had only been 18 months since I had left Apple, which meant I qualified for an employment bridge; my stock options, employee number and previous employment time all rolled into my current employment phase.
This iteration of Apple was more stable; there was no more OS 9 group, the clones were gone, ATG was cleared out, Betrand had a functioning software organization, the product lines were much cleaner, Bas and the UX team were cranking out good designs and Steve was CEO and ruled with an iron fist. It was this organization that produced Tiger; the first release that I felt really represented the vision and aspirations of what a desktop operating system should be.
The funding did occur during the dot-com boom, however the quality of the team was quite high. Mike Boich, Andy Herzfeld, Susan Kare and Bud Tribble were all members of the early Macintosh team. Mike had several successful previous startups return large amounts to the funders and Andy was also well regarded for Radius and General Magic as well.
Working at Eazel was great! I was there when the team was quite small; less than fifteen people. Eventually our tiny workspace was bursting at the seams and all the cubes were dismantled and rows and rows of picnic tables from Costco were brought. OK, that part sucked big-time. It reminds me of high density floor plans at Amazon.
Most of us were on IRC all the time and it was always funny to see a majority of the office silently stand up to go on a lunch trip that had just been arranged on an IRC channel.
Andy was a huge proponent of the Free Software Foundation and Richard Stallman gave us our employee orientation. Yes, he did put on a robe and wear a hard disk platter on his head during the final convocation.
Because of the reputation of the founders and the quality of the staff, there was always someone interesting stopping by; Jeff Raskin, Steve Jobs, Alan Kay, Bill Atkinson, Mitch Kapor, Heidi Roizen, etc. I remember most of these people because I would always have to ask myself "How did a kid who grew up on a farm and didn't have a college degree end up at a place where I could talk to people like this?" I had a similar feeling when I was at Be and Apple as well.
The early GNOME developers who were worked closely with were also an amazing group of people. Nat Friedman and Miguel de Icaza, Raph Levien, Havoc Pennington and many more.
I mentioned the Eazel team having a huge impact on Apple as well. I was able to learn from these people, go to Apple with them and watch them do great work. Darin Adler and John Sullivan, who already were legends to me because of System 7 went on to found the Safari team along with Don Melton, Maciej Stachowiak and Ken Kocienda. John Harper, who was respected for the the sawmill/sawfish window manager, went on to create CoreAnimation and I believe large parts of SwiftUI. Bud Tribble became a senior executive behind many of Apple's technology. The Apple Watch is just one of his projects. Pavel Cisler brought his input from Tracker at Be and Nautilus at Eazel to Finder, where he is still managing the team. There are many more that didn't make the trip to Apple, but ended up at Android and other places doing great things.
As for the day to day work, it was a lot like any other programming job. We used gcc, edited code, compiled, checked things into CVS, managed bugs in Bonsai, automated builds in Tinderbox, etc. When I think about it, you could probably take 2001 me and drop me into 2022 and I wouldn't be all that shocked. Same processes, different tools. I would be super disappointed things hadn't progressed the way that I had envisioned though.
For the record to this day I believe this aqua pill styled interface was way more clean and attractive than the flat design. I've never been fan of flat interfaces, it's the reason I hated Windows 8 onwards interfaces. Skeuomorphism was just perfect!
Same, this feels absolutely dreadful! "Where do I start? Why is there so much stuff screaming at me? What's my first move here? Oh shoot it, I don't want to do this anymore."
> Apple releasing their first phone, which will likely run some kind of Cocoa in it. Good thing you already know how to write applications for it, right?
Awww. It must have been so shocking to MacOS devs when the iPhone was announced and it was web applications or nothing.
I bought my first desktop Mac back in 2004 after owning an iBook 3G since 2002. And as a 19 year old software developer from Germany who finished his apprenticeship, I wanted the best Mac I could afford. I bought a Power Mac G5 - one of the first dual core models with water cooling. Was a lot of money back then and I waited months for it to arrive, but it was a great machine.
Played a lot of World of Warcraft on this thing. :-) And did some Cocoa/Objective-C development. But it was never the big investment that I thought it would be. After the Intel switch in 2006, the Power Mac G5 didn't have much more use for me personally because the new Intel machines were so much better...
Today I'm more into buying the cheaper machines for myself. Should have put the money from the Power Mac G5 into Apple stocks...
My first desktop Mac was an 8500. I bought it when I was a student and took out a loan to buy it at £3500 second hand. Unbelievable how cheap computing power has become in such a relatively short time.
Why would you use Yahoo in 2005? I'm pretty sure Google was already the more popular search engine by then, especially for people looking for developer documentation.
Edit: Best data I can find shows Google was about 35%, Yahoo 30%, and MSN 15%. So I guess it was a toss up if you were using Yahoo or Google, but I seem to recall everyone I knew who was a developer preferred Google because it did a better job finding developer docs.
I used Rhapsody for a couple years on a Thinkpad. It was very, very limited for what I was doing at the time, and the Lighthouse applications I could get running weren't anywhere near what anything else was, by the time I got them.
My work laptop at the time was a G3 running System 8.6, which remains my favorite Apple OS to this day. I still have a G4 with 8.6 stuffed full of everything for the heck of it. And the Thinkpad.
I'm still doing security updates, but you have to self-build, and any new development would be "when I get a round tuit." Some people are doing downstream forks and there are build-it-for-you scripts/Automator packages you can use.
Pretty sure it would: I replaced my old PPC iMac in (IIRC) 2017 and it was already basically impossible to get an updated browser version (let alone most if not all the other apps).
It's just missing the part where half way through, osx and xcode updates and your existing app no longer builds on your machine nor runs on anyone else's.
My ascerbic observations about the platform aside, I love this, both in it's current context, and would have loved it at the time. Thank you!
During this era Apple sent third party official devs a shirt when the new OS released. I still have mine but they are quite worn out from all the wear. Wish they still did that but I appreciate how much less expensive Apple developer accounts are these days.
This makes me more nostalgic for the old Interface Builder than it does Aqua. So much more straightforward to create actions/outlets and create the files than the current mess.
This was really fun to read. Tiger was before my time - does anyone know of good resources on where things went from here? What were some of the first 3rd party native Mac apps?
Just for context, I'm 26, from a post-commie country, have been around computers since birth thanks to my dad. He wasn't technical, he just liked the new tech.
What stinks to me:
- I very much prefer 16:9/16:10 ratios (4:3 begone)
- Skeuomorphism was always very "uncanny valley" for me; I much prefer the Win95/OS 7/etc. designs than skeuomorphic ones, the current flat designs are better but way too saturated and I tend to lose my focus quickly
- I vividly remember how lost and frustrated I was when those old IDEs would launch with multitudes of windows by default - most of them were never used by anyone and everyone would just click through to the main window; being a small child I didn't know what to do or where to start and nobody around me could help - this memory kept me disinterested with programming until I literally went to a programming bootcamp after my finals.
- I really mean it! IMHO user friendliness is over the roof compared to those supposedly golden times.
I think the answer lies with the ‘aesthetic usability effect’, which is a well studied phenomenon. Basically people find visually appealing UI more usable even if it is objectively worse.
Hard disagree on aspect ratios. I grew up on 4:3 which was great for working with text editors and long blocks of code. The switch to 16:9 felt like it was just driven by the movie industry for people who had nothing better to do but use their computer for multimedia consumption.
> I vividly remember how lost and frustrated I was when those old IDEs would launch with multitudes of windows by default...
Tiger and Xcode 2 is quite old and I think is not really the version of macOS many people yearn for. It was the first version many started using (10.0-10.3's graphics were terrible - pinstripes! - and 10.4 was the 'Redmond, Start Your Photocopiers' release which was genuinely exciting. Although I'd used OS X at uni, I bought my first Mac with 10.4 Tiger. Since for many it was the first used, it's what's remembered - but using Tiger this month, I realised that several things I fondly remembered were actually in newer versions.)
The 'best' version is very likely 10.6 Snow Leopard (2009), or possibly Mavericks (2013), the last pre-flat-design OS X. Snow Leopard had a clean, fairly modern UI (so your concerns about multi-window were heard) yet was still joyful.
> I very much prefer 16:9/16:10 ratios (4:3 begone)
The iMac G4, which runs 10.4 Tiger and 10.5, has a 16:10 aspect ratio screen if you buy the 17" screen option. (Source: just bought one, to investigate if it's truly nostalgia or things really were better back then.)
> I much prefer the Win95/OS 7/etc. designs than skeuomorphic ones
Much of OS X was bright and colourful, with pretty graphics, which as you note is very different to Win95/OS 7. However the skeuomorphism wasn't as strong as its reputation is these days. Much of the interaction (say, Cover Flow) is what we'd today call skeuomorphic but really was just a fairly natural way to interact. The real skeuomorphic elements, like the Calendar app using stitched leather, were fairly rare.
> the current flat designs are better but way too saturated and I tend to lose my focus quickly
100% agreed. I personally find it very hard to distinguish elements at a glance in modern macOS.
> IMHO user friendliness is over the roof compared to those supposedly golden times.
I think early 2000s OS X was not as golden as remembered, but mid-2000s to 2012 was extraordinary. OS X really ramped up and improved in those years. Then when they switched to flat design, they lost a lot of usability tweaks along with it. Running current and old OSX/macOS side by side on one desk, as I'm doing, you can clearly see it's the same OS, but today's has much more onscreen, taking more space, but has many small UX indicators missing, and yet despite the amount onscreen the design feels austere and soulless. I find UIs with UX hints built in to their design, and designed for visual beauty, both usable and pleasing for my mind the same way any beautiful object is, and I dearly miss them.
I wonder how your iMac G4's arm is. Seems like the 17"s and up all seem to suffer from stretched springs (the lower weight on the 15"s has preserved them).
funny I was thinking this would demonstrate that Xcode and Interface Builder were easier to grok for a beginner than they are today, but it seems it was just as particular then as it is today.
It depends what you want to do, if you're playing around building apps for a hobby and friends and family and you can target Monterey, you could go with SwiftUI and Apple's official tutorials (it runs on older versions but a lot is missing). If you've got experience in React or any other of the declarative frameworks you'll pick it up quickly, it's actually much nicer than React.
Yeah I was there. Some people may think it's just the norm in the old times. It's not. It's just Apple obsession with being "user friendly", and for some reason, they think Gui is more friendly for programmers than code.
It did work though, the developer drain from MS to Apple around that time was massive. They were giving easy tools to low-skill developers, and command-line access to high-skill developers, a win-win. Whereas MS around that time was busy overcomplicating Visual Studio in their quest to merge web and desktop development for lock-in purposes.
Because Jobs was still in control. As I understand it, this was a constant tension between Ive and Jobs. Of course Ive design eventually dominated after Jobs' death.
It appears to me that they're rolling back Ive's decisions, slowly but surely. There are skeuomorphic icons in Monterey, and I think there's fewer borderless crap.
The one thing that does really grind my gears though — they've replaced all purpose-made icons in toolbars and such with some "universal" ones that completely disregard the existence of the pixel grid. There's literally not a single line in these icons that isn't blurry af.
> When questioned about the lifespan of the iPod and why the functionality won't eventually move into the cell phone, Jobs answers, "I'm going to leave the answer to our actions in the future."
For me it was more like “you have just torrented the Golden Master DVD image and restored it on to your bootable Firewire iPod because you only have a CD-RW drive and nobody has released rips of the six-CD version yet” ;)
https://betawiki.net/wiki/Mac_OS_X_Tiger_build_8A428
e: Siracusa’s review for Ars is still a great read too: https://arstechnica.com/gadgets/2005/04/macosx-10-4/