mkingston

https://www.youtube.com/channel/UCR-DXc1voovS8nhAvccRZhg

Jeff Geerling. Also the owner/author of the site we're discussing in this thread.

You can review a PR commit by commit. But it's not obvious how in the UI. Navigate to the commits tab on the PR page and you can navigate through the commits in the PR and review them one by one. Although it's possible this isn't what you mean.

I was reading the git-bug documentation and found "bridges" to third-party platforms:

https://github.com/git-bug/git-bug/blob/master/doc/usage/thi...

I have not tried it.

Don't be disappointed. That's cool! You've independently discovered something.

There's a cross-platform trash/recycling library you might be interested in: https://github.com/Byron/trash-rs

That's a brief statement which makes me think I'm missing something obvious, but it doesn't seem obvious to me. Would you please expand on that?

I mount the board on some nylon standoffs through a takeaway box lid. The rest of the takeaway box is the enclosure. I drill some holes through the takeaway box for wire ingress and egress.

I use a tupperware container if I'm feeling fancy.

native speakers are excellent at understanding accents when compared to other nations where non-standard accents are not as common

Interesting to read this; I had developed exactly the same hypothesis about French speakers: that they're less accustomed to a variety of accents (with the obvious caveat that it's even more risky to generalise from an anecdote than from data!).

I need this for practice. Perhaps I ought to use text-to-speech and then transcribe the audio back to text and compare.

did you ever internalize the importance of the verbs "to be/is/am/are/been/were" and "to have/has/had" growing up as a native English speaker?

Definitely not to the extent that I'm having to for French! At the moment I don't have a lot of spare time to think (eek) and reflect on this, but regarding those verbs, French definitely has some different usage of them when compared to English including sometimes using "have" where we'd use "be", and vice versa, which is a little jarring at first.

You and I know that, but it's not important or useful to a non-French speaker learning English. To those people, those are merely English words. And indeed, they are English, even if they're French imports. I think even a lot of English speakers don't realise the depth of this: "cul de sac" is a fun example that a lot of people I ask don't realise comes from French (or another language, for that matter).

As someone who is currently learning French and (amateur) teaching English I agree in general with your take on English. The frequency with which I say "yes, it doesn't really make sense to pronounce it that way" is high. English is full of crazy inconsistencies and I feel lucky to have learned it as a native speaker. Here's a nice illustration/example of what you've written: https://en.wikipedia.org/wiki/Ghoti

I really appreciate that French is much more consistent with pronunciation but it's very challenging in some other ways. There are numerous conjugations that sound the same but are spelled differently, meaning you have to infer from context where possible. This is quite challenging as a learner.

Il / elle parle ("he/she speaks")

Ils / elles parlent ("they speak")

Both of the above are pronounced exactly the same. It's not such a problem when reading, but when listening and especially when transcribing this is quite challenging. Also, I haven't taken any time to think about it but so far I have the impression that there are far more conjugations of a given verb in French than in English. I've questioned a little lately whether conjugation gives more than it takes. Context, with the occasional "clue" word, seems like a pretty adequate mechanism. In English for example, we don't differentiate between you (singular) and you (plural). But when it's not clear from context we'll say e.g. "you all". This seems to me easier and more sensible (and I'm correspondingly enamoured with the French gerondif), but I do wonder if it's tricky when coming from languages where there would be a difference between the singular and plural.

Which is all to say: I'm very sympathetic to, and I agree with your reservations about English. And I'm also glad that French isn't the language in its place.

I've spent quite some time in France over the past few years and variously dealt with websites of French organisations during this time. I'm not sure what it is, but French websites seem typically to be frustratingly semi-functional. This story is completely unsurprising to me.

Thanks for the reply. Upon further inspection, it looks like it's `lines`. It might also be `filter` but I haven't quickly come up with a way to determine whether that's the case. I opened an issue:

https://github.com/nushell/nushell/issues/8016

I couldn't apply labels myself, I don't think.

Thank you! I've been using Nushell for a couple of weeks and at this point I don't see myself going back. Modern configuration. Great interactive usage (structured data, mostly!). I thought I'd like it, but more so than expected so far. The one shortcoming I've encountered so far is that some built-in functions seem to buffer all their input before producing any output, meaning I can't e.g. tail + filter logs. A fairly simple example:

    journalctl -xefu systemd-networkd-wait-online.service | lines | filter {|in| $in =~ 'start'}

How would you achieve your example with macros? My inclination would be to do this with regex replace and I can't think how I'd do it with macros. I guess there's some functionality I'm missing? Jump-to-first-upper-case-character?

These are of course, good points; cement probably being more porous than many people perceive. Just a little (hopefully interesting) speculation: adhesion might be less important, or entirely unnecessary if the reinforcing material isn't arranged like rebar. I.e. if it were just a tangle in the concrete.

Could you be more explicit, perhaps with an example? Do you simply mean you prefer function/constructor parameters over DI?

Listening to a podcast[1] a while ago and the interviewee talked about the pervasiveness of the tobacco industry in our culture. At the risk of misrepresenting him: I think he would probably argue that the tobacco industry has manufactured peoples choice to use cigarettes, and should be reined in. One anecdote that stuck with me was something along the lines of "in the 50s everyone seemed to agree that if smoking caused cancer, it would be banned, and yet here we are now".

[1] https://lexfridman.com/robert-proctor/

That's a clever thought. Do you know whether connections could be established outside the tunnel after you disable airplane mode but before the tunnel is re-established?

As I read it, it recommends Phillips for manufacturing due to self centering and perhaps cam out; and Torx and Robertson for manual applications. I guess it draws this particular contrast with "manual" applications to highlight that it in fact means automated manufacturing.

I had two commands

watch -c -n 1 -- systemctl --user --state=failed

watch -c -n 1 -- lfs --color yes -c +inodes_use_percent

I didn't actually look for colour output for systemctl (yeah.. bad debugging, kinda lazy, but it wasn't very important so I didn't give it too much time). I thought I'd try to reproduce for you and thought I should have a quick look at colour output for systemctl, and came up with:

watch -c -n 1 -- SYSTEMD_COLORS=1 systemctl --user --state=failed

which works well.

The lfs example still does not; I suspect it's probably ignoring the color arg depending on its environment (I wonder if watch changes the env?- not to imply it would be wrong to do so if so).

I feel a bit bad for making it look like this was watch's fault, when it probably isn't. I'll dig into the lfs thing later.

Funny viddy came up here today. I created this note about watch in my dotfiles just yesterday (edit: probably not watch's fault! oops..):

# TODO: this is ugly, doesn't print colours, is there a modern version?

I couldn't get watch to print colours and styles, with the -c switch and forcing colour output on the "inner" command.

Viddy just works and I've immediately created an alias to watch, and substituted all my scripted watch usage. The history feature looks interesting and potentially super-useful, and I'll hopefully remember that exists when I have need for such a thing. Thank you, and keep up the great work!

Check out Firefox Relay and the corresponding Firefox extension. You can pre-register some-random-string@mozmail.com, which will forward e-mail to your actual address. If you pay for it, you'll have your-own-subdomain.mozmail.com, which acts as a catch-all, and you can use without pre-registering any-address@your-own-subdomain.mozmail.com.

You can reply from all addresses described above. The FF extension makes pre-registration of random addresses a little easier by putting a button on e-mail form fields.

There isn't convenient integration with password managers, that I know of.

You could be de-anonymised. Particularly by correlating your sign-ups using your-own-subdomain.mozmail.com. Or if the service is hacked. There's some argument that makes this service harder to generate spam with, and therefore less likely to be blocked. That's yet to be seen, I suppose.

If you're willing to use a plugin: easymotion (or, for nvim, hop.nvim)

There are OTP apps for desktop operating systems. Hopefully that helps!

I think the rule is this: if your vehicle is considered roadworthy in one country as evidenced by passing a roadworthiness test, called an MOT test in the UK, Contrôle Technique in France, ITV in Spain (those are the ones I know) it will qualify for insurance. If it has passed a qualifying roadworthiness test, and is insured in one country, that is legally valid throughout Europe. There is in fact a larger agreement regarding insurance and roadworthiness that some nearby countries such as Morocco and Turkey also participate in, I'm afraid I don't recall its name right now.

It occurred to me in the wake of the release of these Pandora papers that mass availability of tax avoidance could be one way of shutting down (almost) any availability thereof. Worded cynically: bring tax avoidance to the middle class and the poor, and it'll be shut down promptly.

I was swayed by the argument against working in pairs, then swayed back by your argument in favour of working in pairs. I think the key part that changed my mind was "two students that get along well with each other". I wonder if you emphasise this in your instructions to students? And if so, how?

It's probably a bit late to bother replying to this, but hey, here goes anyway.

Believe it or not, I wasn't arguing against you. My main point was to agree with you, and to inform: markets exist for lessons. The prices are quite reasonable. Some are free. I hear free lessons are typically lower quality.

I also (indirectly) pointed out that lessons are necessarily different from one another. What I've observed is that some work needs to be done even when plans are obtained from another party. Though, from what I've seen, much less. Purchasing lesson plans is a big time-saver.

Perhaps next time someone complains to you about spending too long lesson planning, you should propose they purchase some lesson plans, and see what they say. To me, that's the interesting question. Do they know these are available? Do they find the cost reasonable? Are they opposed to purchasing these lessons in principle, because they already spend their own money for the benefit of their students?

They might do cost-benefit check on whether they might be way ahead taking a second job, using part of the wages to buy plans, and spend the rest on a vacation.

This statement is completely ludicrous to me. Humans should be given the resources they require to perform their jobs effectively. You might be strictly correct that individual teachers might be better off taking another job and buying these resources. But I think you should direct your ire at the employer of the teachers if you acknowledge that this is a problem. Consider the analogous statement:

[Programmers] might do cost-benefit check on whether they might be way ahead taking a second job, using part of the wages to buy existing libraries, and spend the rest on a vacation.