authorfly

11 ms·

61.
by authorfly 2y ago
The comments above you replied to were saying something different to your interpretation, I believe. They were saying that by going remote (the last 5 years), people haven't formed as many deep connections by legitimate social connecti
62.
by authorfly 2y ago
I don't think there are any more which is a shame, because spaCy was an amazing library and probably the library I most ever enjoyed working with, it truly felt like a craftsmans belt for intelligent text transformation/insight. S
63.
by authorfly 2y ago
It's not really a step change. Saycan demonstrated similar functionality with PaLM 2 years ago (2.5 years first demo of cleaning up coke cans, etc): https://www.youtube.com/watch?v=Ru23eWAQ6_E Caveat - I am way out of
64.
by authorfly 2y ago
Hhhm, no not really unless you are in the bottom 30% of earners in the US. You can get US-like outcomes (or worse) on the NHS with the waiting list, or you can get good health care on time with private health insurance. However, you have to
65.
by authorfly 2y ago
I would suggest you prepare to purchase private health care in the UK given the waiting lists.
66.
by authorfly 2y ago
What wealth tax concerns you? Most of the tax begins at $80k+ and then $110k+ yearly income but not so much wealth from my understanding. PS; The Dutch government may reverse the negative expat changes, especially regarding the special stat
67.
by authorfly 2y ago
pioneer.app in their group-weekly update mode was the closest thing to early indie hackers. Sadly they shut down that mechanism, I don't know why. It was great for weekly reality checks but also UI and headline help. That said, I had a
68.
by authorfly 2y ago
Consider how much info cookies and user intent contain about each user via persistent session tracking. I wonder if we will see ChatGPT include a behind the scenes giant JSON with 2-20k fields about the personal in the chat and have finetu
69.
by authorfly 2y ago
For a bit of alternate perspective, I had a happy life until a night shift worker moved in next to me, and began setting off alarms, cooking and leaving at variable times while experimenting with their sleep. They explained how they could s
70.
by authorfly 2y ago
True, but this pattern was established with TrufulQA a couple of years ago. I disagreed then but it's unfortunately convenient to use this approach for some benchmarks.
71.
by authorfly 2y ago
Your library is quite possibly the best example of effortful, understandable and useful work I have ever seen - principally evidenced by how you keep evolving with the times. I've seen you keep it up to date and even on the edge now fo
72.
by authorfly 2y ago
We don't disagree. A look up table with exact rules is a rules system to me from an NLP/GOFAI perspective. I was aware of how the libraries tend to work because I had often used things like looking up lemmas/word sense/p
73.
by authorfly 2y ago
Yes, but you can control that. You can use set fit, less examples, or SVM or etc depending on how much separation, recall and other aspects matter to you for the task at hand. Sensitivity level to biasing to the dataset is a choice of train
74.
by authorfly 2y ago
Check out the training data. Sentence transformer models training data includes lots of typos and this is desirable. There was debate around training/inferencing with stemmed/postprocessing words for a long time. Typos should mini
75.
by authorfly 2y ago
Stemming is basically rules based on the characters. It came first. This is because most words in most languages follow patterns of affixes/prefixes (e.g. worse/worst, harder/hardest), but not always (good/better/be
76.
by authorfly 2y ago
Yes those models were sensitive to the preprocessing, far more than transformers. However Word2Vec and GloVe were fundamentally different, when used as designed GloVe worked better pretty uniformly.
77.
by authorfly 2y ago
Ok, I understand some of this. I know that often the structure of proteins (especially e.g. with ATP Synthase) helps drive and time limits energy production or energy signalling. Just like how the plugging/unplugging of sodium channels
78.
by authorfly 2y ago
The point is they have as much to gain from encouraging people to believe they actively use "TV scanners" and you can't know when they are about with their vans as saying nothing. Cynically it gets your auntie talking, brings
79.
by authorfly 2y ago
Give me a breakdown of how many 9am 30 minute commutes I'll get reasonably rained on (20%+ precipitation) divided by normalized sun flower hours during the non-winter 3 seasons, please.
80.
by authorfly 2y ago
There was a trend in 2000-2010 to publish blog backlogs as books. For me growing up before widespread availability of a home or school computer I could usually access, yet when hacker culture was proliferating in the US, these were a godsen
81.
by authorfly 2y ago
"Please do not park next to our nondescript White Van" would suite just as much.
82.
by authorfly 2y ago
I absolutely don't doubt you but can you provide some accessible education resources or some sports science/biochemistry papers for the claim around "That limited capacity to consume energy applies to exercise, brain activity
83.
by authorfly 2y ago
This reasoning about the subscription price etc is undermined by the actual prices OpenAI are charging - The price of a model capable of 4o mini level performance used to be 100x higher. Yes, literally 100x. The original "davinci model
84.
by authorfly 2y ago
Japan has stagnated and had debt problems over those 40 years. May I ask you if you saw these as they began to accrue and if you think there is a long term strategy to address them? Seems to me as an alien to the culture, that Japan is uniq
85.
by authorfly 2y ago
The 100 baseline on that graph is the highest attention the term has received, and it correlated with a launch and has since decreased. Google never has problems with first the first few millions for consumer-launched tools. They have probl
86.
by authorfly 2y ago
Oh nevermind, I must have misread your comment (or you updated it) to represent the opposite. I think looking for under-seen/under-resolved niches is useful.
87.
by authorfly 2y ago
Exactly. The unspoken thing about startups/the mom test is that now the majority of startups try to do it that way. Well, 90% of startups still fail. So why do the thing that 90% of the time doesn't work? Really successful startup
88.
by authorfly 2y ago
Agree, agree, agree. This is how you make great margins, work with people who want you to work with them, and you have no fear of the competition increasing CPC etc. Can also be more rewarding. Once you have a decade in startups, you realiz
89.
by authorfly 2y ago
Well, it's in academia, in traditional universities, any way. I think corporates are still thriving. I can say from an academic point of view, I knew 4 PhDs who started in 2018/2019, all 4 got depressed and left the field. Their r
90.
by authorfly 2y ago
"Everybody who does Express, React, or any other popular advanced libraries with TypeScript is using these features. Some things are simply more useful to libraries than line of business code - that's fine. The line of business co

More ›