mrbungie

29 ms·

511.
by mrbungie 4y ago
Yet parent has a point. If Yandex (or DDG or whatever) gets you better results in the same environment as Google, maybe it's not about said environment but more about the player (in this case Google).
512.
by mrbungie 4y ago
I'm all for making Google dance like Satya Nadella said. Even if that path includes stochastic "emotional" parrots.
513.
by mrbungie 4y ago
I'm sure people eventually will want and pay for "unshackled AIs" (just using it as a common acronym, not suggesting that they are actually intelligent).
514.
by mrbungie 4y ago
That phrase could be literally be applied to some people.
515.
by mrbungie 4y ago
This. Similar to "OMG. Look at how this person reacted after I [touched its buttons in some way]".
516.
by mrbungie 4y ago
It's at least entertaining and amusing. Enough to be in the news if you ask me. PS (Shadow edit): I'm passing no judgement on the state of journalism, just saying the way things are and have been for a long time. If you don't
517.
by mrbungie 4y ago
This. Also consider that a human brain that is able to figure stuff from as little as a single datapoint is normally exposed to at least 4 years of massive and socially "directed" multimodal data patterns. As many cases of feral c
518.
by mrbungie 4y ago
I expect these responses to be just that, regurgitation. Case on point: I was playing using MRKL/ReACT like prompts with simple models like vanilla FLAN-T5 and GPT-J. Clearly they didn't work as well as using GPT-3 (text-davinci-0
519.
by mrbungie 4y ago
Ozymandias. There have been large and powerful corporations (with supposedly """infinite""" resources) like the East India Company that eventually transformed into no more than dust and rubble. I think the same
520.
by mrbungie 4y ago
Paying 30 dollars once for a Early Access Pass to an indie game in Steam makes a lot of sense. But paying 200/mo USD for practically being a crash test dummy in real driving conditions, I'm not so sure.
521.
by mrbungie 4y ago
Yeah, maybe I should've said "first-weeks/months" level junior data analyst. But hell, as an analyst I would've paid a lot for a tool that searched intelligently through giant datawarehouses (or whatever consultants
522.
by mrbungie 4y ago
You can easily add more components to start thinking about models/assumptions/etc, like adding interfaces to Data Catalogs (i.e. Glue Data Catalog). As part of a POC I made, I built a similar bot without recursion for debugging an
523.
by mrbungie 4y ago
You're asumming the final number id managed by a sole number. Maybe there a "size" value and a "quantity" value. An example would be a float representing the final magnitude over the base magnitude (i.e. thousands
524.
by mrbungie 4y ago
In this case storage should be mainly used for music. I don't think an SSD is going to make a big difference, assuming Sony's software supports playing from SD cards.
525.
by mrbungie 4y ago
Yeah, good idea, but company architects should make a similar advice about NIH though. A dependency is a dependency, there may be tradeoffs between using third-party software and developing new in-house code, but using "Invented here&q
526.
by mrbungie 4y ago
This. I think comments about the quality of an unrefined output from these kind of generative models completely miss the point. ChatGPT and Stable Diffusion are more impactful when you think about them as a paradigm change in creative workf
527.
by mrbungie 4y ago
I guess parent is refering to cold starts. With enough requests/sec and a good scaling mechanism it shouldn't be a problem though.
528.
by mrbungie 4y ago
Did you have any previous experience running workloads in k8s before? Running the Airflow Helm is pretty straightforward, even with more "complex" use cases like heterogenous pods for different task sizes.
529.
by mrbungie 4y ago
Boxplots don't lie, but they can mislead as any summary statistic, data viz or model can. https://blog.bioturing.com/wp-content/uploads/2018/11/BoxVio... Misleading histograms depend totally on the
530.
by mrbungie 5y ago
You're assuming engineers can't be victims of cargo cult, and we all know that's not globally true as it's been discussed multiple times on HN. Marketing teams know that as well, and they exploit that as much as they can
531.
by mrbungie 5y ago
We're going to need some evidence or at least more development on these ideas.
532.
by mrbungie 5y ago
We currently are using Airflow for ELTs/ETLs to ingest from different Postgres databases to BigQuery/Google Cloud Storage. Airbyte looks sweet for the same task and would free us from a big effort burden, but its Postgres source o
533.
by mrbungie 5y ago
It may be flawed, it may be old, but it clearly continues to be sufficiently expressive and useful. Every time someone tries to replace it, it eventually comes back again. Just look at the recent NoSQL and Big Data craze, and then how we&#x
534.
by mrbungie 5y ago
Not in data processing. There are cases where you have so much data that is better to bring computation to data rather than the other way around.
535.
by mrbungie 5y ago
Not to tell you how to raise your children, but please only teach your kid if you see he likes it. Disclaimer: I was a kid that lived my mother's dreams/hopes for a time and now I dread every second of that period.
536.
by mrbungie 5y ago
I've seen it a lot in Data Science/Analytics environments, as it allows for faster query iterations than trailing commas.
537.
by mrbungie 5y ago
Parent is not talking about a 10TB corpus of software but the corpus/set of all possible software weighing 10TB (or less I guess, specially if you take the lottery ticket hypothesis into consideration). A sibling post shows this, it pr
538.
by mrbungie 5y ago
I think a lot of people would expect some evidence for your first phrase.
539.
by mrbungie 5y ago
To reassign crops for feedstock to human consumption is not as easy as it sounds. If Covid-19 has taught us anything, is that our global food supply chain is a complex beast. Also, just a reminder to those lucky enough to survive this pande
540.
by mrbungie 5y ago
An option, albeit a bit complex would be: Debezium => Kafka => Parquet in Google Cloud Storage => BigQuery external queries.

More ›