Richard Whitcomb
4.1K posts

Richard Whitcomb
@rwhitcomb
Digitizing smell @osmo_labs 👃past: @nvidia, @spotify, @twitter, @bluefinlabs
Boston, MA Katılım Şubat 2009
822 Takip Edilen694 Takipçiler

@squarecog I still segment types of code in my mind between glue code and core code. Maybe 90-95% being glue code. Interesting to have a model classify that outside of norms of directory structure etc.
English

Regarding this and the ensuing discussion, it feels like there's an opportunity here for an ML model that determines if a review is required based on the change's interestingness (for shared learning OR risk).
Erik Bernhardsson@bernhardsson
This might be heresy but: 1. Code reviews are a massive productivity tax with tiny quality benefits 2. They should not be mandated 3. The author should feel free to request a review if they want it 4. If you don't trust your engineers, invest more in CI, or hire better ones
English

@karlrohe Similar, would love to just watch old school Linux folks get things done with their terminal.
English

@pdrmnvd They really sold that exactly at the peak. There was only 6 months or so in 2020 where I wanted to get alcohol delivered.
English

@ivan_bezdomny OpenAI is a bit different in that I am not sure it really cares about their users? Goal being AGI over usefulness now. Maybe this shifts faster with the entire board room saga.
English

Funny as this is... looking at Bard -- for the first time in a while, because of Gemini of course -- you realize the difference between a product shipped by Google (via PMs and committee), and one made by the best new startup (OpenAI).
One is exciting. The other has "features."
Nikolai Yakovenko@ivan_bezdomny
That is different. Apparently Bard gives humorous responses now. Or merged with Grok?
English

@deliprao “Spicy Autocomplete” doesn’t quite have the same ring to it.
English

@xLaszlo @chrisalbon What the Greek latter means depends on the field of science or math you are studying. They are normally used to compact notion but are very hard to expand on without deep study in that field.
English

@rwhitcomb @chrisalbon I don't see why do you expect to understand Maths if you don't know Maths.
Greek letters are a convention just like snake case variables. Secondary to the main meaning.
English

My theory:
There are 1000 coding tutorials for every level of coding expertise, from the basics to common patterns to programming norms.
With math it’s like “here is the floor symbol, here is an equals sign, now math the rest of the fucking owl”

Sasho Nikolov ([email protected])@thesasho
It's absolutely wild to me that anyone finds this python code easier to read than the equation. Does anyone who follows me feel this way? Can you please explain yourselves?
San Francisco, CA 🇺🇸 English

@binarybits 90% of baby devices are like 0.01% reduction in parent stress for 100s of dollars that most parent will happily fork over in a moments notice.
English

@binarybits Wait too long and he have a very fussy baby. Also as a parent often you “think” you hear a cry. Monitor is a cheap device that reduces stress and provides a bit more information.
English

@Chris_Said Aside from longer videos I wouldn’t think of any of these any new features. More coming from monetization teams instead of product teams.
English

@rwhitcomb Deverification of blue checks, premium blue checks, blue check boosting, text-free link previews, subscriptions, view counts, longer videos, changes to DM permissions.
Most of these were probably easy to do, and shipped with little to no user research or A/B testing.
English

@rickasaurus I really don’t enjoy being in the Bing app at all. Looking forward to DALLE-3 being in the OpenAI app.
English





