HNHacker News
TopNewBestAskShowJobs

randomsearch

5,690 karma · joined November 3, 2012

the road goes ever on and on
submissionscomments
randomsearch··on People tricking ChatGPT “like watching an Asimov novel come to life”
How can this interface be useful as a search engine replacement if the answers are often incorrect?

Can we fix it?

Because earlier today it told me that George VI was currently king of England. And I asked it a simple arithmetic question, which it got subtly wrong. And it told my friend there were a handful of primes less than 1000.

Everyone’s talking about it being a Google replacement. What’s the idea? That we train it over time by telling it when things are wrong? Or is the reality that these types of language models will only be useful for generating creative output?

randomsearch··on Migrating from Vue 2 to Svelte
You won't be an expert, but you can be productive.
randomsearch··on Migrating from Vue 2 to Svelte
I would say you can equally learn the core of Vue and React in a day, so that you’re actually building your app before the day has finished.
randomsearch··on Migrating from Vue 2 to Svelte
If you want a serious answer: because frameworks do a lot of work for you.

For example, they will keep state synchronised between your data model and components (and with some handy plugins, your server).

They allow encapsulation and reuse through a well thought out set of abstractions.

They provide functionality like routing and app-wide data stores with transactions.

If you’re just building a webpage no problem, use raw JS. But if you’re building a complex app, let a framework take care of as much incidental work as possible.

Same as with all (good) frameworks and libraries, they boost your productivity.

randomsearch··on Migrating from Vue 2 to Svelte
In what way is being on Vue 2 painful? Using it in production, zero problems, everyone understands it, does everything we need.

Plan to migrate to Vue 3 for the nice to haves and to keep up to date, but Vue 2 is not painful at all.

Considered React but I think Vue will be more productive for the team as a whole in the long term. Vue is more intuitive, better designed, and structured. But totally understand React has a larger ecosystem - we’ll just be building more stuff in house so understanding our own code will be more important than having a larger choice of dependencies.

Did take a look at Svelte, agree with you.

randomsearch··on Surely you're publishing, Mr. Feynman
And what are the mediocre coders to do, if only the best coders get hired!
randomsearch··on Surely you're publishing, Mr. Feynman
No, we said fund _people_ not status. Someone’s status is their job title etc. that should be irrelevant in funding decisions.

Funding projects is a terrible idea that you should keep at a minimum. If you fund projects you will end up with the BS we have now - marketing in the form of grant proposals.

I think a really helpful analogy for how science should be funded is to think of how angels and VCs fund early stage startups - I mean, how they really fund them. Which is: is this a promising area to startup within? Is this a great team? And that is pretty much it.

The downside of this approach I think is that personal biases can creep in and create a lack of diversity. You can counter that with pots of funding that directly target diversity.

Fund people, your assessment of their ability, their team, and their area of interest.

randomsearch··on New GPT-3 model: text-DaVinci-003
It’s worth noting that this doesn’t actually make sense. Why would an extraction team come to your house? What is an extraction team for, in this context? Most generous interpretation is the extraction team would be rescuing the trucks. Otherwise, What is being extracted here? The trucks are not at your house, so why send the team there?

It sounds clever at a superficial glance, but what is impressive is its “turn of phrase” rather than any logical thread in the story, which is a non sequitur.

Don’t get me wrong, it’s still amazing, but it didn’t make me chuckle but rather scratch my head to try to make sense of it.

randomsearch··on Tesla.com/.gitignore
https://xkcd.com/932/

I look forward to meeting the Tesla engineers who work on their core tech and also their webpage.

randomsearch··on Why Twitter didn’t go down: From a real Twitter SRE
Is it essential for Twitter to run their own datacentres?
randomsearch··on Why Twitter didn’t go down: From a real Twitter SRE
My first job out of Uni was a company that ran financial services, logistic fulfilment, and retail. They had at least a dozen large in-house retail brands, several large warehouses with automation that Amazon was probably behind at the time, and they also ran serviced some of the most well known brands in the country - from their website to customer support to logistics. They had delivery networks, networks of direct shipping, vast call centres with thousands of staff. We integrated with the banking system, with all sorts of external companies.

Our total headcount was under 10 000. Most of that was logistics and call centres (taking retail orders). Our IT dept was literally about 200 ish. All our office staff fit within a few floors of a medium sized building. IT was less than a single floor.

I’d say our software, whilst serving fewer users, was completing tasks orders of magnitude more complex than Twitter. We didn’t have much down time, in fact almost zero beyond physical damage to our networks and the once every few years disaster. Our critical support team was a handful of greybeards who spent much of the time playing solitaire. Code reviews were tight and access to production strictly controlled.

What we did have was a very lean culture, carefully managed over a long time. The company was over 100 years old.

I do not understand how, with far more sophisticated tech, companies like Twitter need so many staff. We could have run Twitter with a thousand folks, including sales. From an engineering standpoint a few hundred would do. And most of us weren’t great engineers. There were a few very experienced freelancers, though. We all worked 0900-1730, but we worked 7 full hours every single day, every minute of those 7 hours. No beanbags. Most of that time was sat quietly coding in a very cleverly designed open plan office (no cubicles) that was like a library.

Looking back, I think the key insight was that we were very cost driven and absolutely kept everything as simple as possible. We had all the usual project overrun problems, usually because things turned out to be more complex than previously imagined. But it baffles me when I hear engineering head count at places like Twitter. If you had given us 1000 engineers, we wouldn’t have known what to do with them.

randomsearch··on UK food price inflation hit 16.2% in October – cost of food basics surging
Happen to know a couple of people who have worked with Cummings and Boris.

Cummings seems like a reasonably intelligent guy, if ethically dubious and difficult to work with (putting it mildly). What is shocking is that he looks like a genius when compared to those around him. He has basically read around a bit, thinks for himself, and is open to contrarian ideas. He’s just a reminder of how poor our governance is.

I think the reality is most of our govts are not just incompetent but actively harmful in terms of national interest. Obviously a hard Brexit, which Cameron and Johnson created as a byproduct of their own political schemes, is the greatest symbol of this.

randomsearch··on UK food price inflation hit 16.2% in October – cost of food basics surging
Germany’s economy is diverse, and their political system is robust. These two things are not unrelated.

They are a bit bureaucratic for sure.

randomsearch··on UK food price inflation hit 16.2% in October – cost of food basics surging
Single common thread has to be poor governance. Certainly in politics and the civil service and also, from what I hear, amongst business leaders. They are mostly selected from a very small pool, ie public schools via the class system, so it makes intuitive sense that those chosen would on average be less talented / capable (and even motivated) than other countries. Certainly many of those I know who went to private school were not as competent as their grades or standing would suggest. Add to that a lack of life experience (career politicians, sheltered lives, oxbridge bubble) then it’s hard for them to even know what the real problems are, never mind caring or have a vision for fixing them. Simple examples: the productivity gap, the state of infrastructure, chronic underinvestment in the north at the expense of where the upper class live, etc Fallback has been decades of “managed decline”, which is actually - kicking the can down the road to maintain the status quo for as long as possible before people realise the state of things. Thatcher saw the city and selling off national assets as a quick fix, Blair pumped student debt into the regions to borrow from the future. Combine that with infighting and the Tories’ destruction of industry and local government for class war purposes. I don’t see a way out beyond a revolution, but that will take surely many more decades of decline. Best hope is break up of U.K. perturbs the political sufficiently to open it up to the rest of the country. Things are realistically ok for most people (but not for millions living in poverty and literally relying on food handouts to live), but far more worrying is the trend. It’s plain to see things will get very very bad without wholesale reform and long term vision.
randomsearch··on Trump Will Be Reinstated
And in the meantime, whilst Biden is definitely president?

And if trump fails to become candidate or loses like he did last time?

Doesn’t make sense.

Simplest reason most obvious one: trump brings publicity and engagement for his business.

randomsearch··on Trump Will Be Reinstated
Why didn’t he vote for him at the last election?
randomsearch··on The erasure of women from online pregnancy literature
I feel like that acceptance (and I’ve worked at a few places that have just been amazingly supportive and accommodating, long before it was widely discussed) is somehow vaguely related to the way us nerds don’t care about formal dress. “You do you” is the nerd individualistic way, and that seems to apply to a wide range of things, including lifestyle choices, fashion, being trans, and many other axes. So long as you can do the work, who cares how you choose to present?

It’s a neat thing about nerdland.

randomsearch··on Twitter to employees: all office buildings closed, badge access suspended
His stated thesis is to turn it an “everything app”.

Hardcore engineering may be needed to pivot and get to profitability.

https://www.reuters.com/technology/what-is-an-everything-app...

randomsearch··on 100s of Twitter employees leaving after Musk's 'extremely hardcore' work demands
> Who’s gonna say yes to “extreme hardcore” without “significant pay raise”?

I can see one group of people: those who genuinely care about Twitter and have felt frustrated under previous management. But you have to wonder why they didn't already leave.

randomsearch··on 100s of Twitter employees leaving after Musk's 'extremely hardcore' work demands
_This_ is the real existential threat to Twitter. From the outside, before Musk bought it Twitter already looked like a stagnant company making short-term decisions to boost kpi's without a real mission.

I'm just not sure Elon is the man to fix a business that isn't really going anywhere. It was pedestrian, maybe it was hard to see a vision in that space, and the company culture became one of "maintain".

Even if he sells a vision of the "Everything App" like WeChat, is he going to find enough people at Twitter and enough new recruits to buy into that? Wouldn't you rather go put man on Mars or fight climate change?

If he _does_ manage to turn Twitter into the Western WeChat, it will be one of the most remarkable events in tech history. That said, I'm not even sure we want him working on a social network rather than loftier goals, so perhaps it would be better for humanity if he fails fast here.

randomsearch··on 100s of Twitter employees leaving after Musk's 'extremely hardcore' work demands
This could be meaningless hyperbole without detail. What are those systems, why are they critical, how many staff did they have and how many staff do they have now and what type of staff are they?

They have great reason to be hyperbolic - they've just basically been fired, Elon is accusing Twitter of being an inefficient, lazy, engineering organisation. He has insulted them, they will insult him back and want him to fail.

randomsearch··on 100s of Twitter employees leaving after Musk's 'extremely hardcore' work demands
If he needs more than 900 employees, he can just hire more. He is not fixed at "900 employees", so the premise of this argument is invalid.
randomsearch··on Mighty Retrospective
> I wonder what vision did Suhail pitch to PG to deserve high praise. Because this can't be it

From what I've read, sounds like a thin client machine based on the web. Chromebooks but even thinner, relying on SSR. I think that's a plausible vision that could have driven the upfront costs of hardware down in return for a subscription.

Pithy vision - "a new type of web-centric computer"

Edit: I really liked following Mighty, and it's exactly the kind of speculative startup that you can respect and admire even when it doesn't work out. They didn't do anything wrong, they tried to make a useful valuable product but failed. A worthy endeavour.

randomsearch··on Mighty Retrospective
Probably due to time constraints. B2B takes a long time to get moving, B2C allows fast feedback for rapid iteration early on. B2B requires integrations, lots of tangential work that doesn't build the core product.

They may actually have made the right decision - it got them to "time to pivot" without running out of cash.

randomsearch··on Fred Brooks has died
There are not many universally agreed truths amongst us opinionated hackers, but indeed Brooks has bequeathed us one.
randomsearch··on NASA – Artemis I Liftoff
It’s incredible and the more organisations doing amazing things in space, the better! If SpaceX are focusing on unit economics and repeatable smaller launches, and NASA are still on the big ticket strategy - well, having both can only be a good thing!
randomsearch··on NASA – Artemis I Liftoff
SpaceX seemed very scrappy to me, certainly an upstart, and their costs seem offer of magnitude lower. They also seem to have pushed the boundaries of technology, eg by landing boosters - not just on land but also at sea. I think the accusation that they just copied existing technology is untrue.
randomsearch··on NASA – Artemis I Liftoff
Describing anything SpaceX are already working on as “decades out” seems a bit risky given their rapid successes!
randomsearch··on The positive effect of walking on creative thinking (2014) [pdf]
+1 for Compeed. A miracle bandaid.
randomsearch··on What You (Want to)* Want
The way I think about this is: physical laws lead to the emergence of information processing machines that form big networks together and develop cultural language of ever increasing sophistication. The monorail song is just the production of this large distributed thinking machine.
← PreviousPage 3 of 34Next →