Simon Farnsworth

@landaire I've not yet used cargo crev - but it's one that gets brought up whenever I mention cargo vet, so I thought I'd preempt that. It doesn't appear to have a lot of traction, yet, but it might be usable within a community who trust each other as a way of sharing review burden.

@TrueFarnz Thank you! Both are pretty interesting, I'll have to take a look at that them more in-depth later. From the quick look at crev, it might be time for me to start reviewing some crates. Is there enough of a community using crev already to be low-friction?

Bringing in a dependency allows me to just get shit done at the cost of quick validation. I'm not generally going to look at the code, and neither are you. The same is true of builtins for your language/OS, which can have _worse_ attack surface from trying to please everyone

@landaire Nice post! There's a couple of tools you might want to look at and consider adding to the post: 1. cargo vet, which lets you record dependency reviews in your project. 2. cargo crev, which aims to let you share the burden of reviewing crates with the public.

@p0welly @PedalandPost use their cargo bikes commercially, and might be able to recommend you an insurer.

Can #cycling Twitter recommend anyone who can insure a #cargobike for business use?

@MissIG_Geek I have a nasty gut feeling that for anything where "fairness" matters, there's an equivalent of "Arrow's impossibility theorem"; you cannot have all the things you think you want for things to be "fair".

Maybe true fairness could only come from turning these decisions over to random number generators. Yeah, it’s arbitrary and unpredictable - but it eliminates the problems of social bias, asymmetric power, perverse incentives and human arseholery. That’s as ‘fair’ as it gets 🤷🏽‍♀️

What does ‘fairness’ even mean in allocation of limited resources? Is it: Everyone gets exactly the same opportunity and odds regardless of variables? Some indiciduals sacrificed for statistical improvement at whole-population scales? Those in greatest need get priority?

@jonas__rudloff @_N9199 @FreyaHolmer That's not what you originally claimed, which is why this subthread exists - you started off by saying that treating "i^2 = -1" as an axiom is "not really true". But the point of MT is that all axioms are "true"; it's just that they are unprovable truths.

@TrueFarnz @_N9199 @FreyaHolmer No, munchausen trilemma holds.

whenever people ask for the proof of a mathematical axiom I always get a little sad because I know exactly the frustration they are having and what they are asking but,, it's an axiom ;-; it's the thing we use to prove other things, an axiom can't be "proven" in and of itself

@jonas__rudloff @_N9199 @FreyaHolmer Because you said that Freya's choice of "i^2 = -1" as an axiom was "not really true", since "the complex numbers is just a quotient ring". But the only sense in which that is a reasonable statement is the one in which there *is* a useful hierarchy of axioms, rather than a choice

@TrueFarnz @_N9199 @FreyaHolmer Why do you people act like the peano axioms are foundational? There is no foundation in mathematics other then the ones we choose.

@_N9199 @jonas__rudloff @FreyaHolmer Right, but now you have 8 things to go deeper on - why, for example, is it true that for any pair of naturals x, y, x = y implies y = x? Why is the axiom of induction true? Why is it true that for any n, S(n) is also a natural number?

@TrueFarnz @jonas__rudloff @FreyaHolmer Well, if you go deep enough you run into the Peano axioms, so... You can prove it by appealing to them

@jonas__rudloff @FreyaHolmer But all you've done there is change the axioms; you now have the assumptions that are pulled in by the quotient ring (like the existence of the two binops). If I ask you why we say that, in the complex numbers, (x +jy) * (1 + j0) = x + jy, how do you explain that without axioms?

@FreyaHolmer Again not really true, the complex numbers is just a quotient ring of the polynomial ring over the real numbers, and 'i' is just a synonym for a specific side-class in that quotient ring. In that context we can prove that i^2=-1.

@AusterfieldM @p0welly @staustellsimon @OldhamCouncil Cos, as every soldier knows, if you use the wrong thing and it doesn't work, Sarge will focus his attention on you. And you don't want that…

@TrueFarnz @p0welly @staustellsimon @OldhamCouncil Why use RG-58 when Wet Green String* works just fine... *MIL-STD 01

Hi @OldhamCouncil can I have a copy of the report this is based on? I'd love to know how you are defining cyber attack.

@AusterfieldM @p0welly @staustellsimon @OldhamCouncil RG-58, perhaps? That used to be widely used for computer networking, and is originally from MIL-C-17 military specs for coaxial cables.

@p0welly @staustellsimon @OldhamCouncil Everything will be MIL-SPEC. The specification in question with have fuck all to do with IT, though.

@VModifiedMind I'd used cycle paths all the way to the supermarket. They just felt that they could tell me off for cycling somewhere instead of driving. Some people are just that sort of weird - it's happened once in over 100 trips to that supermarket in the last year, so a rare event.

@TrueFarnz Another interesting event. Also bizarre. Why would they have randomly approached you?! Did they feel you’d gotten in their way or something?

Who are these people who allegedly say this? The claims made by pro cyclists amuse me every day. I couldn’t give a toss if someone cycles or drives, whatever transport works for you, but who are these “motorists” saying this, and why if they really do?!

@cfunk1 @RancidBacon @fasterthanlime People jump from "there's no such thing as monochromatic purple" (true) to "there's no such thing as purple". Same mis-step as "there's no such thing as green black body radiation" => "there's no such thing as green".

@mycoliza @josh_triplett In the X11 protocol, clients get assigned large ranges of IDs at connection. They then assign IDs within their ranges as they see fit; it's an error to assign outside that range. That way, client-assigned IDs are globally unique, but the client still assigns them.

@josh_triplett the tradeoff here is that this approach doesn’t work as well if the protocol wants to allow sharing objects between multiple clients, though. client-assigned IDs means that IDs are only unique per-client. which may be okay, but it also may not, depending on the protocol…

A brilliant innovation from the X11 protocol, worth copying in new protocol designs: Let the client making a "create object" request assign the thing ID; don't assign it in the server. Then, the client can send "create object" together with the first batch of operations. (1/3)

@mycoliza @timClicks I'd go further - the conversation about which criteria are the "right" ones to evaluate a language by is the valuable conversation, since it's the hard part of evaluating a language. Once you agree on criteria, evaluation is relatively simple.

@timClicks i think a conversation about whether those goals and requirements are the right ones is absolutely worth having! some people will certainly want a language with different values and different design constraints, and that’s okay!

my personal opinion is that Rust’s async/await is very close to the best possible design for a language with Rust’s goals and design constraints, but don’t take it from me:

@arjansiddhpura @BowTiedDolphin @MalcolmTeas @levelsio I can fit 80,000 OS2 fibre cores, each carrying a 4 THz wide signal into 1 cm². A 1 cm² antenna will pick up a 10 GHz wide signal with similar signal strength and quality to a single fibre core. There's simply no comparison because of the spatial multiplexing wire offers.

There's some vengeance I will feel when 🦖 dinosaur telecom monopolies with artificially limited slow speeds get absolutely bankrupted once Starlink goes mainstream They won't be able to see what's coming at them It's $100/mo now but I think probably will end up being $50/mo and be 10x-100x faster than what you'll get from your local Vodafone or Verizon Next will be phones switching to it: imagine 5G from space, faster and cheaper than the telecom monopolies give you Might be a good time to start shorting telecom companies

@AlecMuffett And the one time she encountered something bad (just mean, not illegal) via an E2EE platform, she was able to get teachers involved. I'm far more concerned by sender-controlled disappearing messages than E2EE - never guaranteed that she'll know how to stop it going away. 2/2

@AlecMuffett E2EE means my secondary school daughter can share her worries with her aunt and her grandmother without fear of her technical father intercepting those messages. This is a good thing; it means that she talks things over *before* they become big problems. 1/2

It is for the sake of our children that I believe everyone - especially them - needs end to end encryption:

@smudge298 @SurreyRS @PoliceChiefs We do for HGVs, coaches and buses driven as part of your job: gov.uk/driver-cpc-tra… IMO, needs to trickle down further, though.

@SurreyRS @PoliceChiefs The amount of Drivers that can’t determine a roads speed limit without signage is worrying, you just have to look at the attendees of most speed awareness courses to verify this. In most areas of life we have continual professional development, why don’t we do this for driving?

It's day one of the @PoliceChiefs' National Speed Operation. What better way to start than a question for you all. What is the speed limit on this road if you were driving a car or motorcycle? (answer will be given later today) #Speed #Fatal5 #KnowYourLimits

@AlecMuffett @GeorgeOu Note, too, that ADSL is asymmetric because the design model was the telco idea of one video call, one voice call, plus incoming VoD for TV watchers. You were expected to pay business rates if you wanted symmetric speeds.

@GeorgeOu Bro? You're too kind - I'm older than you and was doing infosec while you were in High School. We have asymmetric delivery of bandwidth *yes* because it made some sense in the days of ADSL/TrailBlazers; but now it's mostly an offering issue + consumers don't know what it can do.

I've seen this sort of comment before, and generally it's from someone who can't envision people running their own stuff at home but wanting to access it while they are "out" including at work or doing shopping, etc. If people are provided with network nodes, it will happen.