2,760 karma · joined April 5, 2010
There's a contrast between Pi and the kind of chatbot discussed in the article. When we talk to Pi we don't expect it to do anything for us - it just gives advice and makes suggestions that we can take or leave. The resulting stream of tokens matches our expectations enough to satisfy us.
A chatbot on a company's website however, probably we are talking to that because we want something to happen. "Please fix my account", "my last bill was wrong" etc. As the chatbot isn't integrated with the company's processes it can't actually change the state of the outside world and so talking to it will be a frustrating experience. I wonder if this will improve if/when chatbots get better integrated with systems? Will companies even dare to do this for real?
But IMO his finest work was Blue Jam - the radio comedy not the TV incarnation, hour-long episodes of low-key music and surreal sketches. Absolutely brilliant even today. Archive.org has a copy at https://archive.org/details/chrismorris_bluejam. Best enjoyed late at night.
Trigger warning: basically everything. The BBC would never get away with broadcasting it now.
[1] https://kdp.amazon.com/en_US/help/topic/GHKDSCW2KQ3K4UU4
With secondhand players getting expensive now what are the odds of somebody bringing out a new one? I'd pay good money for something that's backwards compatible, has the same chunky metallic look and feel and comes with a few new features.
This sets a good precedent but we'd need other manufacturers (who I'd guess are more cost-conscious) to follow suit before it changes much.
Surely far more of the world's battery production (by watt-hours or kg, take your pick) goes into cars and grid storage than iPhones?
A more practical option would be oil firing, which was tried in the days of steam and found to work, but didn't save enough money to be worth the effort in a country that still had an efficient supply chain for coal. It could be done again now and would doubtless work, but wouldn't smell or look quite the same. Better than nothing I suppose and if coal really does become unavailable, that's what heritage operations will have to do.
Sixty years ago refurbs would be much of the work going on at any large railway works - taking engines apart, stripping them down to the chassis and reassembling almost a new one in their place. The ingenuity today's preservationists demonstrate in doing this without any of the facilities of a plant is truly a thing to behold.
(Source: some of my family restore steam engines. I once tried to drive one and I was terrible.)
If you'd like to catch a glimpse of one or even book a seat on it, check out the calendar at http://www.railtourinfo.co.uk/tours.html.
DevOps/Python consultant with a 15+ years experience building, hosting and fixing applications.
I work with engineering teams to make their products as scalable and robust as possible. AWS, Kubernetes, load testing, lambdas, CI, deploy systems, monitoring, config management, security audits, dev projects of my own - you name it I've done it.
Recent projects include:
+ Re-hosting a large microservice node.js + RethinkDB web application on Google Cloud. The resulting stack (serving a busy news site) is deployed to GKE with a set of Helm charts and a minimal amount of custom tooling for CDN & image serving. GKE has turned out to be low-maintenance and rock solid; I'll be glad to work with it again.
+ Rewriting a Python big data stack for a bioscience client. Result was a robust, scalable system deployed with Terraform/Docker to spot instances managed by AWS Batch. Benefits included better resilience, higher resource utilisation and vastly reduced cost/processing time.
+ AWS/Kubernetes/Docker deployment to support a large distributed (15 microservices + persistence layer) application. Updated a sprawling Django/Postgres app to fit modern containerised hosting, resulting in a stack that makes great use of k8s' scaling & resiliency features. Templating via Helm for multiple, namespaced stack instances & continuous integration.
+ Another AWS/k8s/Docker stack, this time with a persistence layer comprising Cassandra & Kafka with service discovery through Zookeeper. Prometheus & Grafana for monitoring.
+ Large multithreaded Python 3 application for automated crypto trading. Didn't get rich, still posting on here.
hn-2023-01@mbird.biz / https://uk.linkedin.com/pub/alex-hewson/b6/50a/8b4
DevOps/Python consultant with a 15+ years experience building, hosting and fixing applications.
I work with engineering teams to make their products as scalable and robust as possible. AWS, Kubernetes, load testing, lambdas, CI, deploy systems, monitoring, config management, security audits, dev projects of my own - you name it I've done it.
Recent projects include:
+ Re-hosting a large microservice node.js + RethinkDB web application on Google Cloud. The resulting stack (serving a busy news site) is deployed to GKE with a set of Helm charts and a minimal amount of custom tooling for CDN & image serving. GKE has turned out to be low-maintenance and rock solid; I'll be glad to work with it again.
+ Rewriting a Python big data stack for a bioscience client. Result was a robust, scalable system deployed with Terraform/Docker to spot instances managed by AWS Batch. Benefits included better resilience, higher resource utilisation and vastly reduced cost/processing time.
+ AWS/Kubernetes/Docker deployment to support a large distributed (15 microservices + persistence layer) application. Updated a sprawling Django/Postgres app to fit modern containerised hosting, resulting in a stack that makes great use of k8s' scaling & resiliency features. Templating via Helm for multiple, namespaced stack instances & continuous integration.
+ Another AWS/k8s/Docker stack, this time with a persistence layer comprising Cassandra & Kafka with service discovery through Zookeeper. Prometheus & Grafana for monitoring.
+ Large multithreaded Python 3 application for automated crypto trading. Didn't get rich, still posting on here.
hn-2022-12@mbird.biz / https://uk.linkedin.com/pub/alex-hewson/b6/50a/8b4
[1] https://www.thebureauinvestigates.com/stories/2021-07-28/mon... / https://www.forbes.com/sites/afontevecchia/2012/07/16/hsbc-h...
Statistically it doesn't seem likely that Monzo and its ilk would be so overrepresented merely because their customers are more fazed than others about having their money taken away. Still, if the FCA have published meaningful statistics I could be proved wrong.
https://www.which.co.uk/news/2021/09/why-banks-are-freezing-...
Read the articles and look for similar experiences for other banks. There's definitely a lot more noise about Monzo doing it. Maybe (as another comment suggested) there's a selection bias at work but I assure you it's a real thing, and appallingly handled by Monzo when it happens.
> Because maybe they are being forced to hide behind that?
That misinterprets what I said. If there are rules that say they don't have to be accountable for a misfiring fraud prevention system that hurts real people, where is the incentive to fix it?
That's what's so kafkaesque (and frankly rather violating) about the experience: you literally haven't the faintest idea what you've done wrong and the bank refuses to tell you.