241–250 of 309 posts
Earlier quoted context omitted.
put simpler, you learned what not to build.
Shouldn't companies figure this out before wasting tens of millions in budget + working hours? All I'm reading is that corporations are not taxed enough if they are okay with such opulent waste.
Earlier quoted context omitted.
The reality distortion field is strong around anthropic. Anthropic posts tons of equally bullshit blog posts, written entirely by AI, saying absolutely nothing, to the front page and they consistently average many hundreds of upvotes.
I have to wonder if HN is astroturfed. Anthropic and OpenAI can post about a new model and it gets over to HN instantly and is full of glowing reviews of supposedly real people who supposedly have had prior access to it who supposedly think it's the best model yet.
I real life I meet people who like AI and people who hate it, but nobody who’s on a personal mission to defend Anthropic from anyone that dares to question their hyperbolic marketing.
Earlier quoted context omitted.
Genuinely funny how often I see both "HN is an AI hate site" and "HN is an AI fanboy site". As dang said in one of these threads recently, opinions are just spilt on this!
I show dead and never see a dead fanboy. Just the critics.
Earlier quoted context omitted.
put simpler, you learned what not to build.
Shouldn't companies figure this out before wasting tens of millions in budget + working hours? All I'm reading is that corporations are not taxed enough if they are okay with such opulent waste.
ceo had invested £1 million to build a data analytics platform. "democratising data analytics" in a very specific domain. essentially, competing with someone like databricks in a niche. although they had never heard of databricks before i showed up.
For that million pounds they got a job scheduler written in pure django with a halfway finished react frontend. the whole thing was constantly broken. there were multiple race conditions throughout the product. i joined well after the million pounds was all gone. three years after i joined i had fixed the worst of the problems by rewriting massive swathes of the thing.
i eventually convinced the ceo they'd been doing the wrong thing all this time -- they should focus on analytics + specific domain consultancy services instead of software products.
the major failure was no-one ever moved on from idea V1. they never moved to idea V2. which meant they never got to idea V3. instead, everyone spent a hell of a lot of time talking about how great V1 was going to be, and how they planned to build V1 and what V1 would look like, check out this status update about our progress on V1, check out this mock up on what V1 is going to look like etc. they had an agile consultant come in to tell them how to be more agile. a scrum-master to tell them how to scrum.
3 months after joining was the first time i mentioned apache airflow. they literally could have just stuck a nice frontend on top of it and written a backend data transfer library. job done. very cheap idea V1. unfortunately, the previous team of django developers could only see their trusty django hammer. edit -- and i should add their big £1 million budget too.
multiply the budget by 10x or more. exact same thing at some big corpo. bigger budget = room for more bullshit.
Earlier quoted context omitted.
Shouldn't companies figure this out before wasting tens of millions in budget + working hours? All I'm reading is that corporations are not taxed enough if they are okay with such opulent waste.
No. This is like saying "shouldn't scientists figure this out before wasting tons of money running experiments?".
Waste at all levels. I've worked at insurance companies that spent $100million on development where only 200 customers signed up (estimated to be 20,000 at start). I've worked at telecoms that spent $25 million developing internal tools that no one used. I've worked at big tech where entire teams sole purpose is to control a single widget on a UI page.
This isn't even on the procurement side. Recently left a company where a single org was paying $10,000 a month on licenses when only 12 devs existed. I've seen organizations waste tens of millions of salesforce licenses that no one uses.
I'm sorry but the waste is rampant. SMB's can afford to waste tens of millions of failures, but modern US corporations can because there is no real competition in US markets. Just monopolies abusing each other.
I'm sure scientists would love the chance to have stupid budgets and make stupid things.
So no not the same at all.
Earlier quoted context omitted.
Shouldn't companies figure this out before wasting tens of millions in budget + working hours? All I'm reading is that corporations are not taxed enough if they are okay with such opulent waste.
Story time. small business. less than 30 people. ceo had invested £1 million to build a data analytics platform. "democratising data analytics" in a very specific domain. essentially, competing with someone like databricks in a niche. although they had never heard of databricks before i showed up. For that million pounds they got a job scheduler written in pure django with a halfway finished react frontend. the whole…
I worked at a company that had an $80,000 monthly AWS spend when the total users in question was less than 100,000. The most concurrent users was This obscene waste actually isn't health for society nor the economy.
Earlier quoted context omitted.
Shouldn't companies figure this out before wasting tens of millions in budget + working hours? All I'm reading is that corporations are not taxed enough if they are okay with such opulent waste.
Nobody can perfectly predict the future.
Earlier quoted context omitted.
Story time. small business. less than 30 people. ceo had invested £1 million to build a data analytics platform. "democratising data analytics" in a very specific domain. essentially, competing with someone like databricks in a niche. although they had never heard of databricks before i showed up. For that million pounds they got a job scheduler written in pure django with a halfway finished react frontend. the whole…
Yes, thanks for the story. This is what I was trying to say. The idea that it's completely okay for companies to misallocate billions of dollars across the industry while people are legitimately suffering do to myriad of reasons is just bonkers level of selfishness. I worked at a company that had an $80,000 monthly AWS spend when the total users in question was less than 100,000. The most concurrent users was This ob…
constraints can be useful https://en.wikipedia.org/wiki/Oblique_Strategies
> The idea that it's completely okay for companies to misallocate billions of dollars across the industry while people are legitimately suffering do to myriad of reasons is just bonkers level of selfishness.
Yes. The metaverse bullshit comes to mind. Something no-one wanted or needed, and exorbitant amounts of money spent on it.
Earlier quoted context omitted.
put simpler, you learned what not to build.
It's definitely that, which is very valuable, but it's also the optionality value additionally. You had the option to launch the thing, which you wouldn't have had if you had never worked on it at all. It's notoriously difficult to properly value optionality, but it definitely has value, and often a lot of value.
we should just be straightforward, say "we built the wrong thing" and then ask how we built the right-er thing.
Earlier quoted context omitted.
The reality distortion field is strong around anthropic. Anthropic posts tons of equally bullshit blog posts, written entirely by AI, saying absolutely nothing, to the front page and they consistently average many hundreds of upvotes.
I have to wonder if HN is astroturfed. Anthropic and OpenAI can post about a new model and it gets over to HN instantly and is full of glowing reviews of supposedly real people who supposedly have had prior access to it who supposedly think it's the best model yet.