fifticon

the protocols are going if they are not enforced by a google. google was the "US" that gave you the post-WW2 "peace" on the web. The big players only follow "your" standards when it suits them.

you csn run the model, but someone with vastly bigger resources need to train it.

the standardized web we have had until now, was policed by google so they could harvest ad revenue from us. with no gorilla to encorce such standard, the web will balkanize as it's done before.

we are 54 years old

so does roman mosaics :-)

I see mixed comments, so let me add some praise. I am one of countless, who match his intro-filter: repeatedly hearing 'enlightened' people lament that the vast masses don't "get" lisp and FP, and repeatedly attempting/failing to pick up the red string myself.

background - I am a computer science major with 30+ years experience. I did do a mandatory class of 'implement your own lisp' many eons ago. It just never really 'clicked' for me. I do, by accident, assimilation and lazyness,employ FP style designs in my software. And I guess fp techniques gradually rub off on me from e.g. javascript, lambdas,closures, and map-filter-reduce. in particular, lambdas are useful to me. But I am one of the guys who continue to read the "let me tell you what monads really are", and every time I fall off the bicycle. So, well, I appreciated this 'Xfor 5year olds" :-)

denmark has same rule, for same purpose and effect

I did, but the crux is, that this was governed by them driving competitors off the market, so I didnt have options to choose from. You could argue it's the same mechanism, but to me it is important at what point the cut-off hapoens. If I have multiple options, I would pick the more gadgetry one.

well, I have a parking timer installed in the windshield of my car, and it has a 'universally hailed' beautiful design.

It is an obnoxious hassle to replace the batteries on it, and equally so set the clock on it.

So, should I trust thus beautiful design to not be a hassle in daily use..? I long for an ugly parking timer that is easy to use, but those have been driven out of the market by this beauty.

I'm willing to call them equally bad. Making the world inhospitable through global warning, is pretty high on the list in my view..

It partially used to work, but a lot of the legal underpinnings were dismantled in the late 90's, and parts of it possibly also in the reagan era. Right now, MS is doing again with Edge, what they were already slapped rather harshly for, back in the 90's.

These communication platforms are a source of power and control, there are big interests involved who do not want to let this opportunity for power slip between their fingers.

As part of this, it has been rebranded to WhatsAd

intereatingly, for apps you mention '..they offer less control ..'

Maybe for the user, but for the corporation, they offer more control..

they exist, in my country they are available as alternative to smartphone apps for identity auth. (ie you can choose between android, iphone, and TOTP LCD device.)

I have a question: (which I admit I could probably research myself)

It sounds like is is a sort of 'change-events' pub-subscribe approach, where dependent systems keep their state updated by doing deltas against what they currently store (if I misunderstood this part, the rest here probably is moot).

If that is true, is there anything in the system that guards against if a subscriber fails to ingest an event, for whatever technical reason (bug or systems/execution failure, e.g. network or overload), ie some sort of 'transaction' approach where a subscriber will keep receiving/being due an event, until he has reported back an "ACK I have processed that" ().

I am asking because otherwise it could lead to things like a user account staying open/active, even though an event said it should be deleted.

Some alternative mechanisms with other trade-offs could be things/events like "group X has changed, its current member list is now [...]". In that case, the mirrored group might eventually become consistent on a later update (as opposed to carrying an eternal memory left over by a missed update).

() I am vaguely aware true consistency guarantee is impossible in a distributed system, something about confused generals.

erh, the option with the house, sounds rather expensive? my day-to-day work regards professional building cost estimation software, and I would claim people try to do a lot of work to avoid having to do that. I'm not saying they don't sometimes end up doing it anyway, but in my perspective, the larger the scale, the more this is aggressively avoided as much as possible. Similarly, I encounter a lot of comparisons where "we" software people are told to be either more or less like the building people. What I do see though, is that building people perpetually tend to miss out on a lot of data optimisations / pipelines in the building project flow. They keep talking about wanting to do this, but in practice end up entering a lot of data from scratch multiple times. One of the culprits I see, is that the people who should have shaped the data for this, have no economic motive to do so - "why should we do that, it will only be a problem for X other people at a later stage we are not involved in".

An interesting but flawed article I encountered on lobste.rs.

I agree with and recognise its premise - that SO started to feel hostile and decline, before AI "attacked" it. Maybe the article isn't flawed; maybe it just doesn't match my personal expectations/wishes: I would have expected more details on how the late self/auto-moderation reputation game damaged SO. Instead, the article fades out waxing poetically on how developers long for 'authentic community' where we help each other. Put differently, it appeared to promise it would deliver on the flawed reputation game.

Instead, I'll put in my personal anecdotal experiences of this as an SO user. I am an 'average' SO user, doing a mix of the three 'passive consumption', 'asking questions', and occasionally answering or commenting on questions I have trench-warfare experience on / 'been there too, did this'. My SO experience as the years progressed, is that more and more my attempts at relevant contributions would be flagged/killed/removed on spurious judgements and vague-unclear justifications.

The actions reeked of being motivated by "if I moderate/remove N items, I meet my quota", instead of being grounded in valid concern. And of being based on literal interpretations instead of intent ("TECHNICALLY, it is possible to flag this contribution according to rule #237, and since it earns me a cookie point, I'll do it because nothing stops me").

The original intent behind that entire system was to ensure quality contributions. But the actual effect of it being applied this way, just conditioned me to not be bothered to add contributions, with the feeling that at best I would get hassle to be able to contribute, and that instead I was producing dead-in-the-water cannon-fodder for a reputation-farming moderator.

I had hoped the featured article would shed some more light on this, but it never really dug into the subject, just referred to it as accepted fact.

By the reply count it appears this post was excellent at trolling HN's core audience :-). The author may be right, but his preface "look I'm one of you" was in a very different tone to the rest of his "now shut up and get to work" rant.

I do use the AI tools to some extent, if for no other reason than that they are currently the path of least resistance, and google+friends have lately played themselves out of the game.

He is probably right we should get acquainted with using agents before dismissing it :-).

well, if you are russian, it appears not.

Though my daily work involves plenty of DI, and I see the need for it, I see some unfortunate side-effects, in the behaviours it 'causes'.

- the 'autopilot GPS' problem: Colleagues who basically have no idea how things fit together, because DI connects the dots for them. So they end up with either a vague or no mental model of that happens below the surface.

- the same, but related to scope and costs: Costs: Because they don't touch what is 'built behind the scenes', they get no sense of how costly it is ('every time you do use that thing, it instantiates and throws away a million things'). Scope: Often business logic dictates that the construction hierarchy consists of finely tuned parts: You don't just need an instance of 'Foo', you need a Foo instance that originates from a specific / certain request. And if you then use two Bar's together, where Bar 1 is tied to Foo 1 but Bar 2 is tied to Foo 2, you will get strange spurious errors (think, for example, ORMS and database transactions - the two Foos or Bars may relate to different connections, transactions or cursors.)

One antipattern I have seen (which may actually be an argument FOR DI..), is the 'query everything' service, which incorporates 117 other sub-services. And some of the junior developers just love that service, because "then I can query everything, from a single place!" (yes.. but you just connected to 4 databases with 7 connections, and you are only trying to select a single row from one of them. And again, code with the everything-service becomes quite untestable).

The best part of the article is its advice for triggering the broken dependencies at compile-time, I really hate when I have to go through complicated flow #127 to learn that a dependency is broken.

I have a random idea regarding compromised tokens, which may not hold water. What if you put things like the client's IP address in the token? Then the server can reject (and mark for compromise) as soon as they receive any request from a different ip address? I realise this will also invalidate people who somehow roam between ip addressses, say DHCP/wireless in a larger building.

with the wrong admin access, cloud data can be wiped just as easily :-(

the government that slashes your science budget is the same government that would never feed the homeless, unless it was -to the sharks.

no worries, they will also pass a law requiring the sun to shine at night, so it all works out.

I might deceive myself, but as I recall, I was vary satisfied with how world of warcraft handled environmental music changes. imuse may be a whole other level though.

Interesting how this will stop me from taking a picture with my mobile phone. The amount of effort people will go to, to make people's work more cumbersome. I am not screenshotting for espionage, I am screenshotting to accomplish my job.

as I understand it, populism is when you exploit 'popular' causes to get a mass backing to bring yourself into, and to stay in, power. It is opportunistic, 'goal justifies means' philosophy.

So, you don't really have orinciples or a cause, apart from 'I should be in charge'. It also means voters, if they are capable of judging that, cant really trust you, because you will switch causes whenever it suits you.

In my own country, I could name politicians who always carried the same color as the reigning political movement, and who would switch when the winds change, because their real principle was 'be on the winning side'.

Please don't do this to the planet! Oranges only.

Maybe we could give Trump the northern part of Greenland, as visible on this map? As you can all see, the northern part of Greenland is HUGE, which is why I think it is the best part to give him. The small insignificant SOUTHERN part however, he doesn't have to deal with, we can leave it to the greenlanders (appropriate, given that they are already called greenlanders.) Heck, we might even call the northern part HUGELAND! /s :-)