Weird that Sketch gets the "I'm not a fan of subscription pricing" warning when it's not a subscription. Buying it only gives you a year of updates instead of until the next major feature/version bump, sure, but you're free to stop paying and keep using the version you have.
The actual forced subscriptions like Ulysses and You Need A Budget somehow get a free pass though?
I really like Sketch's model and wish other apps that are going subscription would emulate it. It fixes the biggest complaint I have with subscription pricing which is that if you stop subscribing the app stops working and data produced in that app is now being held hostage.
"When purchasing an annual subscription, you will immediately get a perpetual fallback license for the exact version available at the time (of the purchase)."
You buy version A (for full price)
They upgrade you to version B (hoping you will pay the reduced price)
When version C come out if you haven't paid yet you get reverted to version A
How comfortable are people with subscription pricing for a Mac App? I've been seeing a lot of them are going that way.
To give a developers perspective, I have been contemplating to implement subscription based model into my app to make it more sustainable. Putting in time for development and as well as marketing (to always get more users) is frustrating.
It's a function of the straw that broke the camel's back. What non-life-critical apps or services am I paying for every month? Quite a lot already. Cell service and cable TV & internet are already $350/mo for me. Then you have Netflix, Amazon Prime, Hulu, Sam's Club (Premium!), Google Play, some stupid app my daughter needs for $8/mo, Wolfram Alpha to help her with homework, LastPass, Apple iCloud storage, SpiderOak backup, Google apps for business... and I'm probably forgetting several others.
I understand everyone wants a subscriber, not a customer, but my budget is dying a death from a thousand cuts here. I just can't keep paying for all these things, even when they're only a "few" dollars, every month. It all adds up. This is why people are saying they only pay for a subscription if it REALLY matters to them. The slots are full. If your model requires a monthly payment for something, it must literally change my life, at this point.
Thirty years ago, I had a $20/mo land line, and a TV antenna, and that was it! Think about that! I'm not at all clear that my quality of life is $500/mo better than it was back then.
I wonder if in the future there will be a subscription service which consolidates subscription services and pays our fractions of the subscription fee to each provider a la music streaming.
That'll take a few years to come by. Right now (and in the last few years), many content providers/producers have actually been focusing on having their own streaming services to get more control on their assets and earn a bigger piece of the pie. The users end up paying a lot more separately to each one, and so this cannot be sustained for long.
I see some sort of reverse consolidation happening in the next few years. Netflix, Amazon, Google, Apple and probably Facebook would lead this aggregation, as they have been doing (or trying/planning to do, as the case may be).
What bothers me about subscription pricing is how expensive it invariably turns out to be considering the amount of use I get out of a program. I would much rather purchase a program outright. Furthermore, some subscriptions seem to be intentionally hard to turn off (WSJ news and Adobe products).
Professionally, I use Linux, MacOS, Windows, Emacs, programming languages, and TeX a great deal and they are all extremely powerful and end up being either free or just part of the expense of having a computer. I don't mind paying for IntelliJ, its a great product, and I get regular use out of it and JetBrain's other tools. I also don't mind paying for on-line services: Arq backup, Dropbox, etc. These seem to be worth it.
My problem is that I'm a nerd and I like using software. I'm unhappy using low end software and so I end up buying crazy expensive software that gets very infrequent use: Mathematica, Adobe Photoshop, Adobe Lightroom, Adobe Acrobat Scanner/OCR/X Pro, Adobe InDesign, Adobe Illustrator, QuickBooks, SPSS, Final Cut Pro, Logic Pro, Microsoft Word, Excel, the list goes on and on. It even includes games like World of Warcraft.
It would make more sense if this high end software was priced per days of use not for the number of months that I've owned it. I've owned many of these products since they were installed via 3 1/4 inch floppys. Not all of these are sold as subscriptions but they might as well be because of the way the expensive upgrades work (e.g. Mathematica and Adobe products).
> What bothers me about subscription pricing is how expensive it invariably turns out to be considering the amount of use I get out of a program.
I wonder if the problem is you're undervaluing your software. You're expecting a certain price point but often that's too low for a business to be sustainable.
I don't understand what you mean. You can get the Adobe Creative Suite with Photoshop, Lightroom, some other stuff I don't remember (don't use), for $10/month.
Those are individual prices, but yeah honestly they are pretty high. For a business, it's $30/month/app/license. For a company with one graphic designer using Photoshop+Illustrator, that's $60/month (or $70 to get all the apps). That's already $720+ every year. With 9 million paid subscribers [1] that would be at least $90 million / year, assuming that everyone was on the minimum plan and only had a single license. If they gave every developer a $200k salary, that would be equal to 450 developers. Looking at the statistics about creative cloud revenue [2] it seems like they're earning in excess of $3 billion every year, so splitting that up among the same salaries would be more like 15k developers working on the products (no matter how complex their software is, 15,000 developers seems like a massive stretch for relatively minor updates to their existing platforms). Obviously they have other operating costs and they need some net income, but in my opinion it really should not require such excessive pricing, especially when their software is made to stop working as soon as you stop paying. (I actually looked it up and saw that they do indeed have ~15,000 employees (hopefully most aren't developers), but it's honestly insane to think that they need that many people to work on incremental updates to well-established programs that already do 99% of what's expected of them).
That's the single exception, any other single Creative Suite app costs $30/month.
Even at $120/year, I have serious reservations about putting my photography library in piece of subscription software. I've heard that LR6 is the last standalone licensed version, so I'll be looking into Darktable as an alternative.
The photos themselves aren't locked in: they end up somewhere on your filesystem (that mimics the structure you see in LR).
It's only the metadata and other photo edits that you would lose if you were to stop subscribing. And of course you could export those edits out of LR before you stop your subscription.
It's not much different from another piece of software that you buy outright, but that is not going to be supported after X years. You'll have the same problem: "hey photo_something 2.0 doesn't work on MacOS High Sierra anymore, but our v3.0 does, for a small upgrade fee".
Are you going to hold back on upgrading your OS? For how long? At some point all the delay tactics stop working, and by then there isn't enough of an ecosystem for the "old world" anymore to help you transition out smoothly.
$120/year is really not that much if you are a professional photographer (or digital artist) who needs the bells and whistles that PS/LR provide. If $10/month makes or breaks your business, I have bad news for you.
I personally think that $10/mo for LR/PS is a great deal.
The difference is the "X years" part. I'd rather own software that works until something actually breaks and I can choose if I keep paying for upgrades, rather than software deliberately exploding if I stop writing Adobe a metaphorical check every month.
I bought Lightroom 6 for $79 (upgrade price from LR3) about 26 months ago. It still works fine. Whatever features I'd have gotten with the CC package ($260 and continuing upward forever) are certainly not worth more than tripling the price.
Obviously if it's your profession or a more serious hobby the value proposition changes. But the "everyone pays the same price, even the occasional hobbyist" model is pretty shitty for the occasional hobbyist.
I will never use a subscription app if the computation happens in the client. E.g. Photoshop, Mathematica, games, etc. Not even if the price is low. Note that I will happily pay thousands of dollars in this programs as long as they are not subscription based.
I might use a subscription based app if it's just a frontend for something that happens server side, but even with that the bar is very high. It needs to fundamentally require server side computation that can't be done on the client for whatever reason. For example, the cloud Mathematica thing does not pass the bar.
So far no application has passed the bar.
Unrelated, but another turnoff are apps that require .pkg installers. Drag and drop should be enough for anything except perhaps kernel drivers, and I don't want to install 3rd party kernel drivers anyway. E.g. VMware should use the hypervisor framework rathet than its own kernel drivers.
This. I will pay a subscription for the actual ongoing use of someone else's resources, i.e., their computing capacity, their storage space. I'm not about to pay a subscription for code sitting on my machine. And I'm especially not about to pay a subscription for something I already paid for when it was a one-time payment.
One other thing that's easy to forget: you can't just look at how much it costs and how important it is to your users. You also have to consider competitors and how difficult your product would be to replace. Even if your app is worth $120/year to me, I'm still going to switch to a competitor if they'll sell it for less.
TextExpander is a good example of this. They changed from a $45 license to $40/year subscription. It's a well polished piece of software, and it syncs between Mac and iOS and has an iOS keyboard to facilitate use there. Quality stuff. But when it comes down to it, some people just want simple text snippet replacements on their Mac, and you'd be an idiot to pay a yearly fee for that.
Obviously I don't have TextEpander's customer/sales information, but I assume more than a few users jumped ship to aText's $5 one-time-purchase.
If all you need is text snippet replacements on your Mac, you can just visit the Keyboard pane of the system prefs. They'll sync to your iThings, too, albeit a little wonkily at times (like all iCloud syncing).
I think you could raise the price to $30 at minimum. The hybrid approach is also viable.
As the other poster mentioned, you could add features to justify a "pro"-labeled version... maybe pivot-searching backups, or automatic S3/rsync/etc backup.
On the subscription side, an obvious option is hosted backup service that charges per GB per month (hopefully $5-10/m for the average user).
Subscription pricing requires the app to be so crucial you can‘t live without it.
Photoshop, IntelliJ, MS Office — these apps can charge whatever they want, people just really need them to get their work done.
But if your app is a simple news reader or a markdown editor or an activity viewer — you‘re gonna have a hard time convincing people to pay for it every month.
Honestly, with Google Drive around I can't see how MS Office will remain profitable for long. Sure, it can handle some corner cases Google's product can't handle and it's the standard, but I can't see that staying the case for long.
As for IntelliJ, I really have yet to understand the appeal. What does it do that other IDEs can't? Intellisense type stuff already exists everywhere (Visual Studio, so why use ReSharper?; Emacs; Vim; Sublime Text; Atom; VS Code; ...). Is it really just about being first-to-market? If so, why is it essential?
Seriously? Google docs has far, far less functionality. Have you tried, for example, to insert auto-numbered equations or images? There doesn't seem to be a way to achieve that. That's not a corner case. The spreadsheet also chokes on data that excel can handle. Google docs is fine for maybe 80% of use, but I wouldn't call 20% "corner cases", and to cover them will take way more than 20% the effort of building MS Word.
Google Docs is still making Office non-essential for the average user. 90% of students have been purchasing Office for years in order to essentially perform the basic formatting provided by any rich-text editor purely for the purpose of compatibility.
>Google docs is fine for maybe 80% of use
Which is able to undercut a good portion of Microsoft's market.
Keep in mind the original comment I was replying to:
>Subscription pricing requires the app to be so crucial you can‘t live without it.
MS Office is on its way to becoming redundant to the majority of users. The price is going to be less and less justified in the near future as Office loses its reputation as a crucial piece of software (at least for the average student/employee who doesn't need to perform very specific formatting or perform complex spreadsheet calculations).
Even if the average person doesn't need MS Office for their own work, there will always be this one prof that emails homework assignments in MS Word format, or the accountant that sends you an Excel Workbook to fill, etc.
Yes, if you're a programmer you can probably get away without Office. But for many people who collaborate with others, having a copy of Word / Excel / Powrpoint is still pretty much essential.
(Side note: I hate people who use Google Docs for their presentations. "Oops, I guess I shouldn't have pressed the back button". "Sorry for the delay, the presentation should load any minute now". "Could everyone please turn off Wifi on their phones so the speaker can open his presentation!")
I'm still doing mail merges with Word and Excel. Can Docs/Sheets or Pages/Numbers do this? It's the only reason I have Office installed, and I'd love to delete it!
I hate subscriptions because, as @TheRealDunkirk said in another comment, things start adding up quick and quite expensive. For most people's personal lives, content like video (TV shows, movies) are more important than other kinds of subscriptions, like an app for productivity or workout or anything else. So that's where most of the money tends to flow (wherever people are willing to pay).
If you've been following the standard "sell and charge for upgrade once every few years" model, expect a backlash from loyal users when you move to subscriptions. It's quite simple to get this if you put yourself in the user's shoes and look at things. Wouldn't you feel ripped off or locked in by a subscription? Would you prefer some app that stops working, just like utilities to (like water supply stops if you don't pay the bills)? Many companies who've been in the business still fail to understand this simple point (AgileBits is a good example I can think of in the recent times), but they also change their focus from long time/loyal users to new users who don't really care much about comparing things or using many features. They don't see losing long time/loyal users who complain about this switch as a loss (and probably see it as good riddance).
If you're really anxious to get subscriptions done to get a steadier stream of revenue, my recommendation would be to price the subscription very low and aim for a large number of subscribers or price it quite high and work to get just a low number of subscribers. Both have their pros and cons, but this depends on your product, how valuable it is for users (compared to competing apps), who it's intended for, etc. As a user, my preference, if I do want an app or something, is a very low priced subscription. Maybe a dollar a month or better, a few dollars a year would be better (don't laugh, there are apps and others doing exactly this).
I personally don't like subscription pricing for desktop software. I prefer a fixed price that I can pay up front. I like having control of when I update and if I update.
Not sure what you mean about YNAB "getting away with it" - if you check the forums/reddit etc. you'll find a lot of their users are very unhappy about it.
As an incensed "YNAB Classic" user that will never pay for or use the new YNAB, I'm not sure about that on two fronts. To me it was very surprising how quickly the community followed what seemed like a clear betrayal of promises made, although there were a few loud complaints at first. We haven't seen any real objections with teeth, like an open source clone of YNAB 4.
Also it's not clear how much outrage was over the switch to a subscription model (for which there are pros and cons), or for the switch to a cloud services model and the about-face from the "you own your data" position they held before.
I mean "getting away with it" as this list recommends it with no warning about it being a subscription, contrasted with Sketch being given a hard time (despite not being a subscription)
These days, I no longer have a long list of apps. In fact, I don't even have any language or compiler installed on my Mac.
My apps for development are: NeoVim, Docker, Source Control and Paw.
For Communication and other: Skype, Chrome, Dropbox, Banktivity, Tor, Google Earth and VLC.
In fact, I'm upset that I have more apps than I should.
Why should I have Skype when Messenger can make video calls from the browser. Banktivity could have an online version. Tor should be builtin in Chrome. Or why use Chrome instead of Safari? Apple should release a functioning browser. And Google Earth should run on the browser sometime in the future as JavaScript improves. And why have VLC? QuickTime should be able to run the videos I watch.
That means my list becomes: Dropbox. Maybe I don't even need that! Make Dropbox like an external HardDrive or something. Some integration in macOS. And my list is 0.
The last thing I want is more cluttering. A dashboard? What the hell do you use that for? A photobooth? I'm not 15 years old.
Having lots of apps remind me of how I was 5 years ago. You just want more apps to "feel" good and productive. Sometimes it makes you feel important, busy and technical. It's all B.S. folks and it's bad for you.
This shows not that the author is wrong, but that different people have different preferences. For instance, I wouldn't be without IntelliJ, or a host of associated profilers etc, and would generally avoid using web apps where there's a decent desktop app equivalent; I generally dislike the feel of them. You apparently feel differently. And that's fine.
Web apps still feel somehow unnatural to many, and I agree that they usually are of a lower quality compared to decent desktop apps. But compatibility with linux and specially the ability to separate working context completely using browser profiles makes the web my platform of choice. I use separate profiles for each of my primary projects, and it really helps switching and focusing.
It may be bad for you, but I'd suggest being careful about extrapolating from your personal preferences to the general case. I use about a dozen command line tools on mac regularly, and on top of that maybe a half a dozen gui apps regularly. Plus a browser of course. I'm fairly productive (although these things are hard to measure objectively) and I have no desire to have the bulk of those compressed onto another runtime in the browser.
Minimalism for minimalism's sake is just as detrimental.
Also favoring web apps over native versions isn't reducing complexity you're just pushing it from the OS to the browser.
I really miss a Skype app in Linux, and keep closing the web.skype.com tab all the time. I like apps. I don't like web technology being the one and only way to interact with a computer.
I was not aware of the Online version. Seems smooth to me but I tried to make a video call and then it asked me to "install a plugin". Well, I don't want to install the plugin for the same reason I don't want to install apps. So I guess I'm stuck with the Desktop version for now.
> Or why use Chrome instead of Safari? Apple should release a functioning browser
You're being vague here. Safari is one of the fastest browsers if not the fastest (on mac). Most popular extensions are available for Safari too, so that's never been an issue for me. Apart from the web inspector, I honestly don't feel like Safari is a bad browser at all.
I'm surprised this list doesn't cover Spectacle[0]. This free app definitely surpasses Windows 10's adequate window management, and completely supersedes the embarrassingly poor built-in functionality.
I love Spectacle but use Moom instead because it supports the same features that I used Spectacle for, but it also supports ignoring the gab (when maximizing windows and so on) that is created on the side of the screen the dock is at when the dock is set to auto-hide.
I hate that gap, and Moom allows it to disappear, pretty much.
He's recommending Clean My Mac. I am a bit out of my medium on Mac, even though I use it 8hrs a day, but I was under the impression Clean My Mac is little more than bullshit, albeit with very effective and somewhat shady marketing practices.
Now I feel terrible for bad-mouthing an apparently respected company...
I could swear I've seen those "Your Mac is having a problem" ads that fake being a system error trying to get me into Clean My Mac 3, but may be it was MacKeeper.
I use Windows, but way less than Mac and Linux, so I might not be very up-to-date. I use CCleaner regularly and it saves several gigabytes of disk space every time, do most Windows user not need/desire CCleaner now?
You are almost certainly mistaking it for MacKeeper, which is what pretty much everyone does when I recommend it. It's really sad because CleanMyMac (CMM) is one of the only cleaning tools I've ever used and can recommended, particularly on a mac. It's surprisingly high quality for a clean up utility.
The uninstaller is very convenient for me and I use it often, as it tries to detect configuration files and asks to remove them as well.
For uninstalling applications it is quite useful. By default macOS applications still do not have a good way of defining which files belong to which application and when you simply remove the .app file a lot of crap stays on your computer. It has also handy tools to remove developer cruft one accumulates.
MacPaw is actually quite respected as a developer. Enough so other developers trust them to steward the "Netflix for apps" Setapp.
When you upgrade from one Mac to another, you can use "Migration Assistant" to get all your apps, docs, and settings on the new Mac. This _also_ brings across all the leftover cruft from apps you no longer use, but whose data is still in your ~/Library folder (hidden, contains the working data for most apps)
Repeat a couple of times, and your current Mac will have large amounts of files that are leftover from your current mac, your previous Mac, _and_ even from your earlier Mac!
"Clean My Mac" will help remove this accumulated cruft.
Source: Mac developer whose product is guilty of using ~/Library to store a large database.
They have rather aggressive marketing tactics for sure, but I've installed it the other day, ran it, and it freed up almost 70GB on 512GB driven to 90% capacity. App looks and works professionally.
I use Clean My Mac, mostly for the privacy aspect of removing usage history/etc. It's not strictly necessary (just like CCleaner on Windows is not - your PC won't explode if you don't CClean for years) but could help giving you back a few gigabytes of cache/temp files.
On the repo owner's Pixelmator line:
> probably the best image editor out there on Mac, is packed with very powerful features and is very simple in its UI
I feel like that's quickly being overtaken by Affinity Photo[0]. One time fee, though you have to pay twice if you want a license on Windows and MacOS, but it's a closer approximation of Photoshop so you won't have to change too much muscle memory. It has better non-destructive layer editing as well. I found it crazy hard to do a layer style like desaturating a layer without "baking" it in.
Been using Photoshop and Pixelmator, Affinity Photo definitely is the in my books the best image editor currently. Love that most of the things in Affinity a real time, so you see changes immediately for all the effects etc.
It's still under work of course, and hasn't reached the kind of UI maturity that Photoshop has, but it's getting there, and you can't beat their price either.
I used to be like this... crazy about dotfiles, backgrounds, shortcuts, apps, etc. Eventually, I learned that I'm better off with the standard setup and little customization. No fancy aliases, no recorded dotfiles, no crazy editor configuration, nothing. If I get a new computer or am using someone else's even, it's easy for me to install what I need as I need it and get going out of the box.
>>
Twenty years ago, you expected a phone to be provided everywhere you went, and that phone worked the same everywhere. At a friend's house, or a restaurant, or a hotel, or a pay phone, you could pick up the receiver and make a call. You didn't carry a phone around with you; phones were part of the infrastructure. Computers, well, that was a different story. As laptops came in, people started carrying computers around with them everywhere. The reason was to have the state stored on the computer, not the computer itself. You carry around a computer so you can access its disk.
In summary, it used to be that phones worked without you having to carry them around, but computers only worked if you did carry one around with you. The solution to this inconsistency was to break the way phones worked rather than fix the way computers work.
<<
What I do is the VDI approach. I have one AWS Windows instance with and I RDP to it from any of my computers at home, the office or any of my second homes, which happen to be iMacs but could be anything.
In my case it is indistinguisable from local text editing. I remember however that I chose to turn off SublimeText scrolling animations because it was a way worse than locally. My fiber latency to my AWS instance is about 60ms and with phone 4G about 80ms.
I do not have exact costs at hand because it is just one of several AWS instances we have, but probably my fully loaded costs on a 160 hour month for this instance would be something like:
- 150GB SSD: $15
- 150GB 30 last days of daily snapshshots as a backup: $3.45
- 160 hours of t2.medium (4GB) Windows: $8.64
The typical applications I use are Visual Studio, Delphi, Office, Chrome, and some domain specific apps.
Admittedly, you can reduce further the SSD cost of $15 to 160/744*$15=$3.22, by snapshotting and deleting the SSD volume each time you shut down but I never did that optimization because startup time would then not be seconds but something probably in the range of 5 minutes, as you need to have some lambda funciton create a new instance, create a volume from the last snapshot, shutdown the newly created instance, replace the boot volume and finally start the instance with the right boot volume.
What I also like of this approach is that I do not have to overprovision disk or instances, if I ever need a larger drive, I just modify the volume, if I need a bigger instance, I just shutdown and start with a bigger instance size.
If I do not have fiber or my ISP is having a bad day, using my phone G4 as a personal Hotspot I get some 80 ms. I use USB instead of wifi to save 5ms. To save some money, I shutdown the AWS instance when I do not need it because it starts up in seconds.
For web browsing I often use the local browser because with Macs swiching the current desktop is just a key stroke away (which is not the case for W10 if one of those desktops is the RDP client.
It's not really true, since he wrote his OS, text editor, programming language. Now he just installs the language and the editor/IDE wherever he goes and uses that.
Why not get the best of both worlds? Spend time tinkering to find force multipliers, then throw those dotfiles or Brewfile or installation scripts into a git repo and it's all a `git clone` away.
What works for you certainly works for you, of course; it's just worth noting that there exists a very productive middle ground in which one can use tools without spending an unproductive amount of time on hyperconfiguration.
Yeah. I've started doing this. Warning though, make sure to use a private repo.
I knew someone who accidentally uploaded their AWS credentials to a public repo; that was a huge shitshow.(btw, don't put cress in repos at all, but at least a private repowill give you a buffer)
IMO that's a bad trade-off. You're going to be using your personal computer more than the time you spend switching computers. In that case it makes sense to customize your set up to fit you. How often do you get a new computer?
True for most people, but I have been in situations where I got new Windows machines pretty often. Then I learned to just suppress my itch to customize and just the darn thing.
Now I am my Mac since a few years and also transferred the whole thing once from a TimeMachine backup, so I customize. Though that list of the OP... SO MANY APPS?! I get weak knees just from scanning that list. Though several I may pick, like the Focus app and maybe a few others.
This is the primary reason why I stopped using Windows. The defaults wound me up rotten and after a fresh install it would take me hours (literally) to configure Windows to run the way I liked my workflow. It just got worse and worse with each new version of Windows pushing itself further and further away from my ideal workflow. But with Linux, as much as I have my preferences with desktop environments et al, I found I could more easily work with whatever was put in front of me (after all, if all else fails I can just fallback to Bash). So I gave up fighting Windows and just switched to Linux full time.
This was around 15/20 years ago now and I do have a custom Bash profile and custom tmux config. The tmux config never needs to leave my workstation but the bash profile gets copied onto each server the moment I SSH onto it (as I've aliased SSH to do this) so that means I have a familiar environment on remote systems with zero extra maintenance.
Copying your bash-profile is good as long as soon as you're _the_ user on that server, or when you have separate users for all users of the server.
At my work, we all use the same user for ssh (uid 1000), for all servers, for all projects. It's a good opportunity to exercise mode-switching when switching from (local) fish to (remote) bash and back again, and I'm happy both of fish, bash share the common readline bindings, it also makes it clear what kind of shell you're in if you're jumping back and forth between docs, manpages and shells.
I have one itch, fish and bash do word-boundaries differently when deleting with Ctrl-W (delete word). bash even does word-boundaries different from Ctrl-W when using Meta-D (delete word in front).
I don't know your specific set up so I don't want to come across as preachy but it's not generally good practice to have everyone using the same UID. I know for a fact we would fail our regular PCI and gambling commission audits if we did that but even for businesses that don't need that kind of regulation, sharing a UID means you're sharing passwords / private keys which seems a major security breach just waiting to happen (eg what happens when someone leaves the team?)
You don't need to share private keys or passwords to share a UID, each person can have his or her own public key in .ssh/authorized_keys
Of course I agree that this is still bad practice from an auditing perspective (and assuming a password is needed to su/sudo you do need to share that password).
> Though several I may pick, like the Focus app and maybe a few others.
And when you're used to those apps, you might pick up a few more here and there. And a few more. And in a while you'll have a list just like the one linked.
I used to be a tinkerer, happily maintaining my own fork of, say, Rails. Or Postfix. Or the Linux kernel. It was tremendous fun and a great way to learn internals, but also a colossal time sink.
I don't fight the platform anymore, I just use whatever's vanilla out of the box. And I'm actually a lot more productive as a result.
I think it's more than okay to go through the tinkerer phase, it's one path to growing as a developer.
My Linux distro progression was Linux From Scratch -> Gentoo -> Ubuntu -> Kubuntu -> Ubuntu -> Kubuntu...
My personal laptop has been on Kubuntu LTS since I bought it seven years ago, and it works fine. My employer-provided laptop is a Mac, so I use MacOS there (with some tweaks, especially to get keyboard shortcuts for window management). My home desktop runs Windows 10 for gaming, but I find it fine for occasional dev work.
Basically, I use whatever is in front of me, now. I seem to have gotten more flexible as I've gotten older.
> If I get a new computer or am using someone else's even, it's easy for me to install what I need as I need it and get going out of the box.
Mackup [1] might also be worth considering. It symlinks all of your config files to a supported storage location (Dropbox, Google Drive, iCloud, Git etc.) which enables you to either backup the settings or sync them between one or more other Macs. To restore your settings on a brand new Mac simply run "mackup restore".
Yeah this guy is pretty crazy with the automation and keyboard shortcut apps, I mean 3 different apps for automating keyboard shortcuts or other automation ?
I mean it's a good idea, but I prefer simplifying my own workflow so I can perform it without any external apps. I used like the idea of automating setting up a workspace, but now I like just to open up and close whatever I need at that moment, it really goes into your muscle memory after a while.
Only thing I really customize is my programming environment, my vim setup is highly customized, also shell functions. But each to his own :)
Nowadays on Windows I do a few customization, like showing file extensions, enabling a separate process for each explorer Window and a few other things and that's about it.
On GNU/Linux, I have long settled on whatever is the default WM for the respective distribution.
It's a Mac. Installing an application doesn't necessarily mean "configuring and tinkering" like on Linux. You drag it and it's installed. Apps have predictable GUIs, so they are often easy to use right from the start.
I don't spend time "configuring and tinkering", and yet after several years on a Mac I use a good number of apps.
Agreed. I prefer being more minimalistic. For example, he uses VS Code, a separate markdown editor, a separate markdown previewer, and a LaTeX editor - but VS Code has a markdown mode built-in and can show markdown previews, and I'm pretty sure there are LaTeX plugins for it too.
Well, you collect more and more tools to simplify your life and solve repetitive tasks during years of usage. Sooner or later it's going to be a looong list of utilities.
I used to configure everything five years ago, I installed Linux from scratch and customized EVERYTHING. But then I got tired of it and now I hardly customize anything. I use a note keeping webapp that I created for platform interoperability. No more desktop apps if I can avoid.
Mmmmm, my main thought reading the list was mainly 'thats a huge amount of cruft to overcomplicate my life'. I mean, there are a few things in there that are useful.
On the topic of macOS power usage: I made a free open source app that lets you use the side buttons on your third-party mice for system-wide navigation[1], just like in Windows. Other apps can do this too, but practically all of them bind the buttons to annoying keyboard shortcuts and frequently exhibit unexpected behavior. Mine is (sort of) event based and works a lot better, including in Xcode!
Not to toot my own horn (I mean the app is pretty simple) but it's a splinter that's been bothering me about macOS for years and years.
Starting to go off topic but have you checked out BetterTouchTool[1]? It's not _just_ another window tiling app for the Mac. It also lets you conveniently remap trackpad gestures and taps, 3rd party mice (yes you can remap your side buttons if you want), keyboard shortcuts, etc. The coolest thing is that if you really want, you can also restrict your remaps to work only in certain applications. I've been using it for years and just wanted to share.
I really wish something like this were available on Linux. I moved an old Mac to Linux (no OS updates from Apple), and the biggest things I miss are good support for multitouch gestures with Apple Magic Trackpad and an app like BetterTouchTool that helps make the standard experience even better. I bought BetterTouchTool (when it went to a paid app from donationware), and I would buy something like it if it were available on Linux. I spent quite sometime researching for solutions and trying some stuff, but right now the experience of using the Magic Trackpad on Linux is just the same as using a mouse.
For the specific purpose of remapping side mouse buttons, BTT only allows you to bind them to keyboard shortcuts. As far as I can tell, no tool except for the one I made lets you bind them to virtual swipe events (which generally work better).
BetterTouchTool has always had the "mimicking standard gestures" predefined actions which really send gesture events and don't just send shortcuts. However it currently only supports three finger swipe left and right (which are the most useful to go back and forward). I probably should look into how to synthesize other gestures as well (there is no official API for synthesizing gestures but afaik people have reverse engineered how to do it, e.g. https://github.com/calftrail/Touch/tree/master/TouchSynthesi...)
For three finger swipes I'm using a trick to synthesize them without private API, but for other swipe types I'd need to synthesize them using the private API... However there haven't been too many requests for other event types thus it's not very high up on my TODO list.
(I'm the author of BTT)
Also BTT allows you to bind them to many many predefined actions in addition to keyboard shortcuts, e.g. "Trigger Menubar Menu Item" which can be very powerful.
Huh, neat! I was wrong, didn't realize BTT supported that. I'm using fake three-finger swipes for my app as well, via the calftrail code you mentioned. (Hence the GPL license.) Are you using a different trick? Is there a more elegant way to do it barring private API access?
I'm basically using CGEventCreateData to save an original system three finger swipe, then I'm just generating an event based on that saved data and refresh the timestamp and mouse location before sending it.
Thanks for the info! I was considering doing something similar when I was building my app, but calftrail's code saved the day...
Wish developers were legitimately able to generate those three-finger swipe events, because they are ubiquitously supported and really useful. (Just the other day, I discovered that Preview allows you to switch pages by swiping up and down! Weird.)
Although BlockBlock is in "beta," it works well, preventing installation of persistent items unless given a user okay but running silently in the background. It's fun to see what causes the warning window to pop up, like whenever Adobe Flash gets auto-updated. It's saved me from at least one very sketchy install.
I'm still surprised all these tools are free, he could easily sell them for real money. I've started supporting him on patreon because it felt weird to not pay anything for it...
Useful as many of these utilities may be, I'm worried about the fact that I have to fully trust each and every one of them. There are 63 entries on that list. How am I ever going to be sure that _all_ of them are safe?
This concern is starting to really affect my use of software more generally. I have found many useful browser extensions but I rarely install any of them because of what they have access to.
Open source or not doesn't make a whole lot of difference either, because I'm never going to be able to review and compile all of it myself after every single update.
The somewhat surprising consequence is that the built-in features of operating systems and browsers have become much more important to me than they have ever been.
Essentially, the software I use is
(a) Built into the OS or browser
(b) Coming from one of a handful of organisations I trust
(c) Purely Web based
This is on the desktop. And on mobile the "solution" is to severely restrict what software can do and give disproportionate power to some gatekeeper who will then predictably abuse that position by extracting a 30% cut from everybody and impose content restrictions way beyond what can be justified by computer security.
For me, a massive advantage of Alfred is the clipboard history. I know there are plenty of other clipboard managers, but I've tried them all, and Alfred's works the best for me.
Features I like:
* Ability to set upper-bound on individual size
* Ability to ignore clipboard CF_ types
* Ability to retain variable amounts of data by time
* Ignore certain apps' clipboards
* Filterable history search
* Short and long content previews
I also use iTerm, but that's mainly for the performance benefits and customizability - I could live without it vs Terminal.app.
I felt the same about iTerm until I started using it, and still feel the same about Alfred (have it installed but reverted to using Spotlight instead).
iTerm's great for the features it has that the regular terminal doesn't. Fullscreen borderless, universal hotkey to reveal, more configurations for workspaces, and I much prefer the paneling/tabs setup with iTerm too.
I assume the same argument is used for why Alfred is good, but I've never really gotten into it. Anyone mind trying to sell me?
The extensibility of Alfred makes it a whole order of magnitude better than Spotlight. Check out the Alfred ecosystem over at packal.org. The Spotify and Github workflows are my most used shortcuts.
What Spotlight makes up for by being the out-of-the-box default option, it lacks in versatility due to Apple's walled-garden policy.
Aside from the awesome feature that lets you point to any folder for the config file (I keep it symlinked to my dotfiles), iTerm2 can go borderless, which I absolutely love (iterm2-nightly needed)! http://i.imgur.com/GENVtUJ.png
I think people switch over from Terminal to iTerm too quickly, without giving it a good shot. After using iTerm for a few years I switched back, since I realized it wasn't bringing me any value.
The main advantage that I get from iTerm is the Hotkey window, which allows me to open a terminal window that fills the top 40% of my screen by pressing CMD-`, and then hide it by pressing the same key. Because of it, I wind up being majorly more productive because I feel like it's easy to open up a terminal window instead of doing things the slow way in Finder.
It's fun because directly above the Textexpander comment is Day One, and directly below it are 1Password and Ulysses (all 4 apps have almost same subscription model).
AgileBits has dark patterns all over on its website that people wouldn't even know about this by default unless they searched or someone else told them. As a business, AgileBits has appeared as shady as some other unpopular names, IMO.
For how long? The writing is in the wall. Will next version support it, after it has given "ample time" to move to the subscription model? This version is already under a time-delay popup.
I'm not sure yet -- still using the non-subscription/non-Cloud 1password. I probably have it easier than most, as I don't like the browser plugins and 1password-mini, and I just use 1password as a store for passwords -- I open it, search, and paste when I need to.
So I can just go to using an encrypted DMG + plain text if push comes to shove.
I used Wallet by Acrylic before 1password, so I might return to that.
I've been using KeePass on Mac with an app called KeePassX for nearly a decade. The user experience is not as slick as an app like iPassword, but it is free and secure.
I can't live without it ever since I first tried it. I really hope Apple either buys them or sherlocks them (preferably the former). Either way, I want the functionality integrated into MacOS.
I can't figure out if it's just poor implementation or just terrible UX design, but every time I scroll on that site a sign-up dialog pops up. I can click out of it, but when I scroll again, it immediately pops up. I'm unable to scroll without this stupid dialog showing. (Chrome 60, MacOS)
I closed the tab after about five attempts to remove the dialog. I still don't know what the product does.
Clicking 'No thanks' stops it from re-appearing for me. Yeah, bad design. The app is still great. It lets you take screenshots that float on top of your desktop (like stickies), and can be resized and annotated.
There are a few points I repeat while they're still true. I like Signal for its reputation and standing on security, but it's highly deficient when compared to Telegram.
Telegram provides a much richer UI/UX than Signal or even Wire. [1] It has multi-device sync and multi-OS support. Signal lacks multi-device sync. Signal does not even have a proper desktop app. Signal explicitly prohibits backing up the data and restoring it if you move to a new phone/device (at least on iOS). The backup cannot be done to iCloud or even a local iTunes backup. So don't buy a new device ever if you like your chat logs. Or take screenshots of the chats for reference whenever you do. To me, this doesn't make any sense whatsoever. People want usability a lot more, and if a "super secure" app is not really being very useful, it won't get very popular. I'm still waiting for Signal to get ahead so I can switch to it, but every time I think of it, Telegram looks a few years ahead of Signal.
I believe Telegram owes its popularity to its rich feature set and first class treatment of its desktop clients (Qt5 + Cocoa).
It doesn't help that not everybody uses Chrome as their primary browser (I don't), making it somewhat silly to keep open for a single thing. Spinning Signal Desktop out into an Electron wrapper would be preferable but is still far beyond ideal...
> The same hacker who previously sold data dumps from MySpace, Tumblr, LinkedIn, and Fling.com, is now selling more than 100 Million VK.com records for just 1 Bitcoin (approx. US$580).
Ah, those were the days, when bitcoin was worth <$1,000.
Nobody mentioned Stow[1]. Simple and straight forward.
"GNU Stow is a symlink farm manager which takes distinct packages of software and/or data located in separate directories on the filesystem, and makes them appear to be installed in the same place."
I'm curious about one thing. I recently discovered the wonderful world of a physical notebook for keeping track of everything with my life. One of the things that I love the most about it is that I can leave the computer, zoom out from the technical nature of my problems and solve them in a more conceptual manner (while taking a break from screens).
For me personally one big advantage that mind maps have is that they are digital. So I can access any single mind map in few keystrokes by searching for the file in Alfred. I am also pretty fast with my keyboard so prototyping ideas and new concepts is really fast for me.
I still use notebooks for sketching things but digital mind maps have too many advantages to dismiss. I also recently made an Alfred workflow that allows me to essentially query any of the digital mind maps I made and present all of the contents of these maps in Alfred. Here is the workflow :
Great collection. My favourite editor for notes/code is Quiver. It does Markdown, WYSIWIG, code (via ACE editor), MathJax and even markup for diagrams. It has a cell based approach so you can mix and match different sections if you don't want everything as one big Markdown block. The only downside is that there is no iOS editor, only viewer.
I like Quiver but I wish it would store regular Markdown files instead of JSON. Equally important an Android app that can add notes or alternatively Simplenote syncing would make me consider switching from nvAlt.
The list is a little outdated for current Mac users. For example Karabiner Elements is what I use on my Mac although all features aren't supported yet in comparison to Karabiner. As side note although I love my Mac I find it to support backwards compatibility in apps much worse then Windows. Many of my XP apps still run in Windows 10 although I wouldn't necessarily encourage you to use them.
I miss Karabiner, and regretted upgrading to Sierra as soon as I found out Karabiner wasn't supported. I miss the ability to map both Escape key and control key to capslock key, which is essentially the greatest thing for my Vim productivity.
Pixave looks great, and is something I've been looking for, but I wonder if it's still alive. There's not been any activity on the developer's (previously very active) Twitter account since March, the demo version is several versions behind, and support@littlehj.com bounces. I mention this as a concern because it seems a bit buggy.
Thanks for sharing this. I found it very helpful. I haven't used karabiner before and I'm trying to understand how you're using karabiner but I'm having trouble doing so. Do you ever plan on writing something explaining your usage or do you have any resources that you recommend?
Is there a Vim equivalent of Snippets lab? I recently switched to using vim-wiki which I think could be used in this way to store snippets, but I do not know yet how to configure quick search.
Stay away fom SnippetsLab. It looses data very often... better approach: Use a decent editor like VS Code and use Dropbox/Google Drive with a directory structure of your own.
Great list. I hear you re subscriptions but people who are new to (I.e. don’t own) many of the apps you mention...
Paw, iStat Menus, Ulysses, Gemini, 2Do and quite a few more
Can get fully working copies on Setapp for a really low monthly price. Setapp is a heavily curated subscription app bundle from the MacPaw folks. People cd save a bunch of money and also have access to a large number of useful apps that once in a while are ideal to do a job but that you wouldn’t purchase for one-off use.
It's interesting you ask. iCal had the option of 2-, 3-, or 4-day view, which I liked a lot. Also, I think iCal did much better job visually distinguishing the current day. But I'm also talking about the task management aspects that had been integrated into iCal. Sure, there's Reminders, but besides being a separate app, it is buggy (crashes a lot) and the ui/ux is nowhere close to what iCal offered.
Telegram you don't need to share your phone number, easily synchronises between all of your devices, stores large files, has better controls for managing group chats.
There's always the network effect: if 90% of the people you want to talk to are on Telegram vs about 20% of them being on Whatsapp, then Telegram is vastly preferable.
Also stickers. Custom stickers that are easy enough to set up that lots of people do.
Pros:
- Secret Chat
- Share different file types up to 1.5 GB
- Multi-device access
- Supergroups and public channels
- Telegram Bots
- Lock chats and Ability to hide last seen for particular contact
- Edit Messages and Mention People
Cons I can thin of:
- No calls support or ability to backup chats
- lacks user base
I know, and I use and like Telegram. But the result of optional end-to-end encryption is that most people don't use it. The whole idea of WhatsApp and Signal is to encrypt everything end-to-end to defeat mass surveillance.
Honest question - do you consider only surveillance done by governments as mass surveillance or would you add "surveillance" done by Facebook (to better target ads) also into that classification? The reason I ask is because metadata from WhatsApp (whom you messaged, when, from where, etc.) is used by Facebook unless you opted out (or if it's been declared illegal in the jurisdiction you live).
Signal is the only one to trust for security, privacy and handling mass surveillance better than the other apps.
Whether it's done by a government or Facebook, mass surveillance is still mass surveillance. But at least, with WhatsApp and Signal, only the metadata are observable, but not the content of my messages. It's still an issue, but something I'm willing to accept in exchange for a communication tool I can use with my non-technical friends, family, co-workers, etc.
But you do need a phone number to setup Telegram. I like the fact that it allows not having to share one's number with others by setting a username and sharing that instead (though setting a username increases the chances of getting spam/unsolicited messages especially if your profile pic looks like a human female and your name sounds like one too).
Wire [1] on the other hand, doesn't need a phone number and can work with an email address.
The actual forced subscriptions like Ulysses and You Need A Budget somehow get a free pass though?