HNHacker News
TopNewBestAskShowJobs

kolektiv

1,625 karma · joined December 25, 2008

London/Horsham (and anywhere else interesting).

Tech Theory + Practice. I care most about changing the world in meaningful ways through technology. I care very little about yet another X.

Contact me at andrew at xyncro dot com.

irc:kolektiv:irc

[ my public key: https://keybase.io/kolektiv; my proof: https://keybase.io/kolektiv/sigs/0d5gliRyPmqTUqjf5C4XNh5y1sry6und7d3z4LYWCyo ]

submissionscomments
kolektiv··on ‘Lolita’ Escaped Obscenity Laws and Cancel Culture
Sigh. Nobody has cancelled Dr. Seuss. The publisher has decided not to publish a small number of his works which contain some stereotypes and characterisations which are pretty far from what most people would see as "fine" now. Nobody has made them do this, nobody is going round collecting and burning copies already in print. Many books stop being published all the time for all kinds of reasons - being out of step with contemporary mores and experiences is, and always has been, one of them.
kolektiv··on Why AWS loves Rust, and how we’d like to help
And as a secondary thought around Rusoto (which is awesome, but could be even more so) - it would be great if we came up with some way of covering the non-HTTP/REST type services - for example, there are various streaming services (think speech to text, text to speech, etc.) which are simply not covered by Rusoto. So you can configure those services using Rusoto, but not use them. I've written a few medium-hacky clients for things like AWS Transcribe, but it would be great to... well, not have to.

Rusoto is an awesome project though, huge thanks to those who've got it to where it is.

kolektiv··on TailwindCSS v2.0
Ok, so I could mark everything as "brand-dark" and then change the definition of that in the CSS? Mmm. Ok. How about layout and sizing? A lot of that looks a lot more defined - I'll pick a random example from their own homepage then: "border-b border-gray-200 py-6 flex items-center justify-between mb-16 sm:mb-20 -mx-4 px-4 sm:mx-0 sm:px-0" - if I wanted to change my approach to centring, justification, and sizing, it still seems like a lot of search and replace...

(I think I understand that the approach is that I'd then change what those definitions mean centrally - but that won't help me if I decide that two properties which are currently different should be the same, or vice versa).

kolektiv··on TailwindCSS v2.0
I don't do front end dev, and haven't for a very long time, but can someone help me understand how this doesn't lead to inflexible hard to maintain soup? Just to take colours, for a moment. Ordinarily I might define semantic classes, let's say "primary-nav" or "prominent-action" and apply them to relevant things (and use relative selectors for things inside them, etc.). So if I want to change the colour of all prominent-actions across my web app, I just do it in one place. This seems to be demo-ing adding a particular colour class to everything I want to be that colour. So... If I change my mind, what now? I have to find all those places and change the applied class? Search and replace?

This seems like the literal opposite of the clean separation I always understood to be the point of styling being separate to markup...

kolektiv··on Single Page Applications using Rust
Well, that's definitely a sweet-spot, but as a by-product of that, Rust is still a well-thought through language with a strong and logical type system, some very useful features for expressing domain models, etc. and a model of memory management which maps very well to WASM. So if you already know Rust, this is a short and logical step. If you don't already know Rust, I wouldn't learn it just to write web apps in - but if you do, you now have all the advantages of "same language client/server side" as well as the advantages of safety, good libraries in certain spaces, etc.
kolektiv··on Dorking: the use of search engines to find very specific data
The -ing part of a place name is Saxon, from memory, something like "ingas" meaning "people of". So, People of Dork, literally, and whatever became Dork (a person/place/event) is probably lost to historical memory. We have an awful lot of places with "ing" in them around here, probably due to a lot of C7 Saxon settlement.
kolektiv··on Dorking: the use of search engines to find very specific data
Almost due south of Dorking, down the road in Horsham. The small town was the first thing I thought of on seeing this title. I'd imagine this area actually has a fair few HNers, as it's in the tech catchment area for London, Reading, etc.
kolektiv··on Understanding Serde
Serde is very strong indeed (although not the easiest to understand, this would have been useful when I implemented a Serde library for [de]serialising with the AWS DynamoDB AttributeValue format).

That said, I've always had a slight problem with this approach to serialisation in general, which is that it makes it very hard to have multiple serialisation approaches for a single type (e.g. I want one casing rule for one target, and a different one for another). This seems effectively impossible with an approach based on annotations/macros. Would love to know how others have dealt with that.

kolektiv··on In Defense of Richard Stallman
Sigh. No.

It's not just about one incidence of some mistakenly pedantic comments, as has been made clear by many, many people. It's a pattern of behaviour over years combined with no interest in listening to other people or accepting that he might be in the wrong. His behaviour towards women in particular has resulted in a lot of people sharing stories of his behaviour which, on their own, should be enough for someone to be considered a problem. This latest incident is just what it took to finally tip things over the edge.

This article then goes on to talk about people like Stallman being needed, or essential for something like free software. It doesn't matter, even if it was true, which there is no way of proving, it's pure assertion. Phrases like "required someone who couldn’t take a hint" might sound all perseverance and dedication to this writer, but perhaps going and asking women what comes to mind when they hear about people like this might be an eye opener.

Without him students wouldn't have as many free tools? Maybe, but maybe they would have other tools, better tools, built by the people his behaviour, and the behaviour of people like him, have driven from the industry before they got a chance to have the impact that people like him seem to assume is a right.

kolektiv··on Await in Rust
I spent quite a lot of time writing Typescript using Fluture and a few other libraries a while ago. Coming from a functional background, combinators + futures actually feels quite natural to me, so I looked at the examples of the "bad" code here and thought I actually quite liked it. But I certainly don't mind the async syntax too much either. Will see how it works out some time!
kolektiv··on A Type of Programming
Ok, I see you've added some text in terms of "Buy your copy to read more" which is a start (although it's probably too far from the email capture for people to associate them, and it doesn't stand out from the rest of the body text). But if you want to be transparent, include the pricing there. And if I'm really going to think it's transparent, don't ask for my email without a) really needing it (i.e. as part of a transaction I've signalled clear intent towards) and b) making it clear how it will be used (in normal sized type!)

For me, the bottom of that page could be something in essence like "Buy the book now for a 'pay what you want' price and get immediate access to X content." I'm not sure you'd need much more (email can come as part of that process, and I'd ditch the bitcoin only payments for certain levels - just simplify it and if you need the bottom price to be slightly more to cover card fees, just increase the price to cover them).

Put it all front and centre - if it's a commercial book, that's fine - don't try and hide it, show the value it has and make it very easy to give you money.

kolektiv··on Building a Stateless API Proxy
Just a quick supportive comment on this - it's a really nice piece of writing, introducing a topic which tends to cause people to glaze over, and doing it in a very approachable and easy to pick up way - and that takes some skill and effort!
kolektiv··on Building a Stateless API Proxy
Yes, you could potentially do something like that. A cache is only in essence "I keep a local store, if someone asks for something which is already in the local store, I give them that - if not, I go and get it, put it in the local store for the next person that wants it, and then give them that". Of course, having things expire (knowing when to remove them from the cache) is not a trivial problem, but there's a lot of prior art out there (look for "cache expiry") and there's also quite a lot built in to HTTP to ensure that HTTP content can be cached well (proxies are a very common thing - look in to "HTTP cache headers").

Of course, in your case, it might be debatable in terms of utility - if you're trying to replace a scan, you probably are trying to get results as "up to date" as you can - which would be prevented if the cache avoided doing that! The cache would have the same rate limits, and so you would be just as well off adjusting the frequency you currently scan. Of course, if you can't control that frequency (perhaps multiple uncoordinated scanners) a proxy is one way to give you that control - so maybe useful!

kolektiv··on A Type of Programming
Yup, also annoyed by the dark pattern. To add insult to injury, below $20, it removes the option to pay with a card, and only allows bitcoin - interesting approach to tax, perhaps! Whatever the reason, I don't use bitcoin, and even if I wasn't already annoyed by the bait and switch, I wouldn't be bothered jumping through so many hoops. What a shame, the content looked as if it might be good, but no chance of me giving money with that level of disrespect for the reader.
kolektiv··on The Burden of Proof: Why People Must Support Their Arguments
I... get where you're coming from here, but I'm not sure I'm going to agree. Sure, yes, to some extent you can go all the way to Descartes and believe nothing but your own existence. But back to those works in C minor - there's a lot of evidence of them. Recordings. Sheet music. Orchestras who claim to have performed them, and so on. Now sure, you can say that all of those things are also assertions and not facts but at some point the onus flips, and it's going to be on you to provide some reason for why those things should be disregarded.

If you're not going to operate on a basis of "reasonable doubt until falsified by a better theory" then you're where? In an entirely post-fact state, where sure, everything's an opinion. And sure, if you want to be there, nobody has a right to stop you but it's not a very productive state.

Yes, journalists tend to be pretty careful with claims around things where there is reasonable doubt. But there is a reason journalists are not still writing about Mozart and using careful phrasing like "and it is also believed by some sources that other composers may have composed comparable works in the same key". For humanity to work we have to go with "fact until shown not to be" otherwise we wouldn't make it through a day.

kolektiv··on The Burden of Proof: Why People Must Support Their Arguments
It's not really a difference of opinion though is it? Whether the day is nice or not is a matter of opinion. Whether this is a tasty meal is a matter of opinion. Whether anyone else wrote works in C minor is a matter of fact - or not, and he's provided evidence that shows that other people did - fact established.

This is the whole problem occurring these days - too many things are seen as "well, it's a valid opinion". No, if it's claiming something which can be falsified, it's not an opinion it's a mistake (charitably - if it's done with less than wholesome intention it's something else - a lie).

kolektiv··on British Doctors May Soon Prescribe Art, Music, Dance, Singing Lessons
So, as a UK resident, I'm torn here. On the one hand, many people would absolutely benefit from these things. It is appalling in our society how many people suffer from something as ostensibly simple as severe loneliness, which can cause significant mental and physical health problems. So yes, this sounds great.

On the flip side... Mental health treatment in the UK is currently dramatically under-funded and under-resourced. Waiting lists are stretched beyond belief, and obtaining treatment is almost impossible for certain conditions. (First hand experience - one fairly significant mental health condition, but - ah! Only one! We only have enough resources to even try and help people with two conditions - co-morbidity. Come back when you're worse!) So two things - how much of this will in reality be pushed in an effort to avoid prescribing more serious therapies that people genuinely need (but the system can't manage), and how much of this would actually be available? It's all very well telling someone with little to no means through ill health to go and take up dance, or learn an instrument, but is that really on offer? Will that really be funded and available in areas where currently there's an 18 month waiting list to see a simple counsellor?

(I thought I should pre-empt a possible reaction of "well, that's what you get with socialist/communist/etc. medicine - <death panels> etc." - there's no inherent reason for this other than poor management, under-funding and structural incompetence. It is definitely possible for the NHS to work well again, but right now, the fact that it doesn't is more political than theoretical. It would be like me pointing to the current shutdown in the US as evidence that governments aren't possible.)

kolektiv··on The Eno notation language
From a relatively short glance, I don't see any mention of what problem this is trying to solve. Data interchange? Storage? Consumption/production?

I see some excitingly magical looking type "inference" happening though. That ruby code is sure that the value associated with Malaga is a lat/long pair is it? Not a temperature range, etc.? So, some very optimistic approach to types, no real way to validate content, and some quirky looking syntax (<key>: <value>, but also #<section>\n<key>: <value> apparently? If I have more than two levels?)

I'm not sure what this is for. I am pretty sure that those people using it are going to run up against a whole thrilling raft of edge cases...

kolektiv··on Yahoo to Spin Off Its Core Businesses
Hey, it's ok, you can both be wrong :)

1979/80 would both put you in Gen X as a birth year ('61-'81, although Harvard for some reason thinks '64-'84 - either way, bad luck, Coupland novel it is).

kolektiv··on Resigning from the Haskell Platform
It doesn't have that flaw! But that's probably partly because it doesn't have typeclasses, so Foldable/Traversable aren't possible in the same way...
kolektiv··on Why i’m done wearing a helmet
I'm not saying they're the same thing - I'm pointing out that as risk increases, you add safety features. Drivers in certain situations (racing, for example) do wear helmets, as the risk is now judged to be sufficient. A 10mph ride around the block may well be safe - but the 10mph part isn't so relevant - it's the idiot in a car doing 60mph who doesn't look for cyclists that is probably the significant part of the risk. So...
kolektiv··on Why i’m done wearing a helmet
Yes, madness. Given expected likelihood/severity of potential accidents, one seems more worthwhile than the other. In situations where the risks are raised for some reason, you will find that people do wear helmets. I will never, for example, do 100mph in a kart without a helmet (plus various other safety additions).
kolektiv··on Why i’m done wearing a helmet
In which case yes, just a very misinformed original author. Shattering, cracking, etc. are all known, planned features of a helmet impact - anything which can absorb energy in the helmet and not the head is a good thing (with some obvious caveats of course).
kolektiv··on Why i’m done wearing a helmet
Also, (can't read the article but assuming some kind of context), even if the styro compressed rather than shattering - that's still energy being used for something which isn't cracking your skull. It still seems like it would be beneficial.

If the author is claiming that once it's compressed it won't be as good next time, then that's insane - helmets aren't supposed to be re-used after accidents (whether bike, motorbike, or car racing - I've had to sadly replace my motor racing helmet after an incident, and it's a lot more than buying a new bike lid - but you still do it without flinching because it's your brain on the line here).

kolektiv··on Pony – High Performance Actor Programming
Not the only meaning though - people in the UK (especially some parts) may have different reaction: http://www.urbandictionary.com/define.php?term=Pony&defid=92... :)
kolektiv··on Introducing F# 4.0
I hope you don't get downvoted, those seems like reasonable questions. I'm part of the F# community I guess, so I suppose I could try and answer!

I'm not sure the F# community is the same as the broader .NET community. I think at the moment, with functional programming being a little more niche still, it's perhaps quite self-selecting. It has a lot of people in who really care. I think that's a good thing - I hope - but obviously the broader .NET community is perhaps more heterogenous - it can feel quite enterprise-y or business-y in places, unsurprisingly.

I hope the F# community wouldn't be found lacking, unfriendly, etc. We certainly try hard to be welcoming and inclusive, and I think generally do OK - although we could do better I'm sure, as could every community. There probably isn't as much open source in the .NET world as there is for Ruby/Node as much as anything else because of the backgrounds of the languages and communities (Microsoft has a strong influence over certain parts) but there is still quite a lot, with some very enthusiastic smart folk.

What's stopping people adopting it? Probably too long an answer for here. I would say that I think there are fewer and fewer things preventing adoption as time goes on, especially with F#, and there are some cool startups trying to do big things.

I'd like to think it's friendly, and I'd also like to think it's always open to ideas from other communities. If you wanted to chat as Ruby or Node dev (or anything else) I hope you'd be welcomed.

(Final note - NuGet is kind of like NPM yes - it's a common binary packaging form for dependency management. It has flaws, but is commonly used, and is certainly better than nothing. In the F# world at least, Paket is doing very well - it's compatible with NuGet packages, but perhaps slightly more accurate about using and managing them!)

kolektiv··on Introducing F# 4.0
http://fsharp.org/use/mac/ is probably a good place to start - if it doesn't work for you for any reason, do please complain - the community is really active in trying to make everything better! (The general section on cross-platform development may also be helpful - perhaps more when you've got a little way in!)

Additionally, there's an F# room on http://fpchat.com/ (a Slack group dedicated to functional programming) and you'd be more than welcome to come and ask questions there too - I recommend that venue in particular as it's very active at the moment.

Hopefully that's enough to have a play - reply here if I can help at all too.

kolektiv··on URI Templates (2012)
Oh yes, exactly! That was the note I was referring to obliquely. They're right, in a technical sense, but "wrong" (subjectively in some cases) when it comes to designing tools that work with URI Templates... It's absolutely a smart enough bunch of folks to get the tradeoffs, I just don't envy having to decide them. Even if, on this occasion, I may have argued more.

(Of course, a lot of standards, and this is one, are reflections, and unifications, of existing implicit standards. People had already started using some of these features, so they couldn't really drop them, and still see adoption. Not an easy balance!)

kolektiv··on URI Templates (2012)
Hah, and in the week after I wrote parsers, renderers etc. for URI Templates... Anyway, they're cool, but I wish to <deity of choice> that they were actually reversible. They so nearly are, but they buggered it up with Level 4 modifiers. For example...

    /{lista:*,listb:*}
    +
    "lista": [ "one", "two" ], "listb": [ "three", "four" ]
    =
    /one,two,three,four
Now, reverse it. You can't because there's no way of knowing which elements belong in which list. It's so annoyingly close to being a pure dual, but it's not. So I've had to make up some irksome rules for the edge cases like that which make me sad, but make matching using templates feasible (wise? maybe not, see note in spec, but that note wouldn't be even needed if only... sigh). URI Templates are a good example of why I wish I worked on IETF specs, while also being glad I don't...
kolektiv··on TempleOS is applying to Y Combinator. Partners desired, send an email
Impressive, that took three days last time.

[Apologies in advance, I have read the FAQ, I am aware of the meta-culture regarding one-liners, but I failed to restrain myself anyway. My profuse apologies to anyone upset or offended.]

← PreviousPage 5 of 11Next →