avnigo

Worth it? Absolutely, without a doubt. Should you? Only if you find it fun, especially if you prefer the keyboard to the mouse, although you can definitely use the mouse in vim too.

I would go with neovim over vim nowadays, and if you want something more familiar that will get you up and running more quickly, try AstroNvim [0]. Later on, if you also like to tinker, you can set it all up just the way you like it.

I've heard of vim being called a PDE (personalized DE) [1], and I think that makes a lot of sense.

[0]: https://astronvim.github.io/

[1]: https://www.youtube.com/watch?v=QMVIJhC9Veg

Standard contracts make a lot of sense to my non-lawyer mind, almost like having the different kinds of open source licenses for software that have become recognizable.

Every time I get those notifications now, I have to remember to close down the application, and go run `snap refresh`. There's literally no other way to make sure the update won't quit my application other than hope it's not running when the update will be applied in 2 weeks.

A better—although arguably still horrible—notification would be to tell you when the application is actually scheduled to be updated so you know when to have it closed. Telling you that you need to have it closed at an indeterminate time 2 weeks from now or else fear "disruption" is clearly not good enough.

Who thought this was an acceptable process for a user to follow and not an utter dealbreaker? Should we still call this snap growing pains?

Updates should always only be applied when the application is starting. If there's a new update, a notification would just tell me to restart the application to have the update applied.

Yeah, in any migration—if you can—it's good practice to run both simultaneously for a while until you're convinced you've checked everything and you're ready to drop the old for the new without much downtime.

I am too, but the Matrix implementation feels quite complex with many moving parts, when apps like Briar, Berty, or keet.io [0] show up and look more straightforward to use. I don't know if that will be its downfall or ultimately actually make it more robust, but I'm rooting for it.

As convenient as it'd be for one implementation to crush everything else and become better because of it, I think different choices are good in terms of having alternatives and competing for features and niches.

[0]: https://keet.io/

I honestly never quite understood why prescriptions in the US need to be literally filled, when in the EU, for example, you can just walk up with your prescription to any pharmacy and buy a prepackaged blister pack of the medicine you need.

Not only do blister packs save so much waiting time — you never have to wait for it unless it's out of stock — but it also feels like there's less chance for anything to go wrong when manually handled by the pharmacist, factoring cross-contamination or incorrect filling, even though the pharmacists would either way be liable for it.

Blister packs are verifiably sealed by the manufacturer, you're more likely to tell if they've been tampered with, you don't have to worry about moisture seeping into the excipient and interacting with the active ingredients, and they provide a visual representation of how many pills you have already taken, without needing to count pills in a bottle.

Congratulations, Michael, thank you for sharing your story!

A cautionary note on using Paypal as your sole payment system. If this is a considerable source of your income, I would caution against only using Paypal or keeping a large amount of money in that account. There are too many Paypal horror stories out there to mention of people getting their accounts limited or shut down without recourse because their automated checks flagged accounts for suspected fraud or violating their ToS. Best of luck!

A map lambda example is what I had in mind when reading the article. I'm not a big fan of the temporary variables, though.

Admittedly the example below is not a perfect solution, but that's where I thought the article was heading when splitting that code over multiple lines for readability.

  map(
      lambda x: x.split('=')[1],
      (url
          .split('?')[1]
          .split('&')[-3:]
      )
  )

Is this still too unreadable or more messy?

> people making the decisions to buy it

Most people just buy a laptop and expect to get Windows on it because that's what they know, unless they know they want macOS and go with a MacBook. Otherwise, if you want a Linux machine, you'd have to hope the manufacturer gives you the option of shipping without Windows, or end up buying it and then formatting. As long as that's the default, Windows won't die, and I would think Microsoft will like to do anything to keep it the default.

When I buy a computer, I want Windows to be opt-in, not opt-out. Ideally, computers should come with a network installer like Raspberry Pi started doing [0], giving you the option at first boot to choose what operating system you want.

[0]: https://www.raspberrypi.com/news/network-install-beta-test-y...

Are you sending each other links or just mentioning the ads in text?

If this is just in text—and I'm definitely not defending Meta here—could it also be that the ads you see have got us so figured out already? The topic you choose to talk about may be influenced or seeded by your environment (online/offline), and one thing leads to the other almost deterministically.

Here's an experiment: try rolling a die a few times or using a random number generator to pick one word or more from a list like the EFF wordlists [0], and then talk about that exclusively.

[0]: https://www.eff.org/deeplinks/2016/07/new-wordlists-random-p...

I guess it's a form of lifestyle inflation. Under limited resources, you tend to be more careful with overheads.

> We will become reliant on a small group of trained translators and experts to report what history—including the documents and papers of our own families—was about.

I think we know this won't be true based on AI getting better and better at handwriting recognition, despite the downsides of offloading this work to machines (e.g. if trained in closed system, could result in censorship and potentially limited validation at the endpoint).

However, I do see a lot of upsides to using computation for such efforts: increases the scale of processed documents by orders of magnitude and enables indexing, search, categorization, and further analyses which would otherwise take many man-months for trained translator and experts to complete.

You can also do that by prefacing an aliased command with "command".

Aliasing "cp" to "cp -i" also prevents you from overwriting files without confirmation. It's more of a preference thing than a recommendation, but overwriting on copy is something that bit me in the past when working a bit mindlessly. I guess the proper recommendation would be "don't operate the commandline mindlessly"?

Matplotlib is definitely a monolith, and it does seem like a huge fight to fix the style, but what is great is you can easily use some of its several predefined styles [0], define your own style sheets or customize specific style parameters through rcParams [1], or install third-party style sheets [2], so in the end the styling does seem much more manageable.

[0]: https://matplotlib.org/stable/gallery/style_sheets/style_she...

[1]: https://matplotlib.org/stable/tutorials/introductory/customi...

[2]: https://pythonawesome.com/matplotlib-style-for-scientific-pu...

Check out the hashrate for Ethereum Classic [0], it doubled overnight. Would be interesting to see if it's sustained.

[0]: https://www.coinwarz.com/mining/ethereum-classic/hashrate-ch...

I was fully expecting to be disappointed and turn it off after a few seconds, but I can tell you, I was transfixed.

I would agree on the tone of the latest keynote, but I wouldn't say it's entirely a departure from their 1984 Macintosh commercial:

https://www.youtube.com/watch?v=VtvjbmoDx-I

The iPhone keyboard actually does something similar by dynamically expanding tap zones:

https://www.youtube.com/watch?v=wPmVKyhyl9U

Another question is what percentage of those is hosted on AWS?

The context of the quote is found here [0]:

> In addition to making sure that location information and other data are fully encrypted, participants’ identities remain private from each other and from Apple.

I don't know that it matters whether or not there's any authentication information. If you could get your Apple device to misreport its location (would need jailbreaking for iOS probably)[1], maybe you could fool the Find My network into reporting the AirTag is elsewhere.

[0]: https://support.apple.com/guide/security/find-my-security-se...

[1]: This would likely need to be quite involved, considering location may be determined from a combination of IP, Bluetooth, WiFi, GPS, etc., although I'd be surprised if there aren't already tools available to do that, especially on macOS.

I can't not share this excellent video on the subject by CGP Grey on "metric paper":

https://www.youtube.com/watch?v=pUF5esTscZI

Not to mention that your email is tied to your online identity in many ways when mostly every account you make asks for an email to use as your username, verify your account, or recover access to your lost account.

Handing over control of your online identity like that to a centralized third party when you could be cut off at any point [insert here any recent news about people wrongfully losing access to their Gmail account and not getting it reinstated] seems like the wrong solution to be allowed the privilege of sending email.

I guess the very least we could do to keep some control is to own (read: rent) your email domain you could move elsewhere in case you lose access, but then you gotta make sure you don't also lose access to your email domain.

Is there an actual solution to all of this?

Here's a fascinating related Numberphile video with Professor Brian Butterworth on the way our brains see numbers:

https://www.youtube.com/watch?v=D1sPBCxlDQQ

It's not just companies selling your email address to third parties, it's also companies that keep your email on file and then get compromised.

Most of the spam/phishing I get is from companies that stored my personal details and then got hacked.

I would say it's likely you just got lucky.

I agree that working on a project is the best way to learn programming, but I have to emphasize: it's the structured learning that ties it all together and allows you to get a better understanding on how to tackle such projects, whether self-learning or guided. You need to be methodically exposed to available tools, solutions, and common practices you may not have come across when working on a project.

I had a feeling this was gonna be about 3D Touch just from seeing the title.

It was definitely one of my favorite features, and Force Touch isn't a drop-in replacement.

With 3D Touch anywhere on the keyboard, you could apply pressure to start moving the caret (text cursor) around, but what was even more useful was being able to apply a little bit more pressure to start selecting text with incredible precision.

With Force Touch, you can long press on the space bar till you get haptic feedback, then you can move the caret around freely. However, if you want to select text, you need to use a second finger on the keyboard after you start moving the caret around, but honestly it's an absolute mess to use; try it out, I can guarantee you won't be able to select what you were planning on.

However, as much as I miss it every time I fail to accurately select text on iOS, I don't mourn its loss. Force Touch is adequate. It's just not a replacement, and seems a shame it was probably removed to reduce costs and hardware complexity, all at the cost of regression.

I'm glad you found something that works for you, and I don't mean to dissuade you even if I could, but to me that feels like an antipattern if you only use it for typed text.

Consider that with a text editor like Vim, for example, you can "time travel" [0] through your file's edits, or even have undo branches/trees [1][2] available per file. That saves you the trouble of having to transcribe text from screenshots, and also barely uses any storage space.

Plain text is also highly more portable and more likely to be recoverable in case of drive failure or file corruption.

Additionally, or alternatively, you could try any sort of manual versioning system or background automatic backup solution that keeps versions of files as you work on them.

[0]: https://vimtricks.com/p/vimtrick-time-travel-in-vim/

[1]: https://neovim.io/doc/user/undo.html#undo-tree

[2]: https://github.com/simnalamburt/vim-mundo

This reminds me of BabyX [0][1], not sure on how different DeepMind's implementation is.

[0]: https://www.soulmachines.com/resources/research/baby-x/

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

That sort of exists, but only sort of. If you press the lock button on the side of the iPhone five consecutive times, it will then require your passcode to unlock (hopefully a high entropy passphrase), and will disable biometric authentication until unlocked with a passcode. You can set the phone to wipe after 10 failed attempts to unlock.

You can also say "Hey Siri, whose phone is this?" and your phone will lock down the same way as described above.

Of course, this doesn't protect from the $5 wrench attack, but plausible deniability only goes so far as well in a targeted attack. At least, depending on your local laws, law enforcement may not be able to compel you to provide your passphrase, but they can easily force you to use your biometric data, so this protects against that.