sackfield

8 ms·

31.
by sackfield 2y ago
As an engineer you really have a finite amount of good working years, and accepting startup salary vs big tech compensation is a bigger risk than founders are willing to generally admit.
32.
by sackfield 2y ago
We should feel great that they now have a competent board that will make sure of that.
33.
by sackfield 2y ago
It is. Like I said above I would be concerned about the continuing success of the organisation according to its operating principles. Destroying the company over childish disputes would not be fulfilling that charter. I'm very glad Ope
34.
by sackfield 2y ago
I'm not sure OpenAI compares to Enron or FTX in terms of unethical behaviour or financial fraud. The only similarities might be, funnily enough, disfunctional boards (or in FTX's case, no board).
35.
by sackfield 2y ago
I would be thrilled, if I were board member I would want the organisation to succeed and ChatGPT can't be described as anything other than an unmitigated success. I wouldn't feel the need to make myself the main character of the s
36.
by sackfield 2y ago
Every culture at some stage has its dark ages, western countries did exactly the same thing not more than 200 years ago. They changed, all things change. Your argument regarding the high water mark of cultural morality could be used by Euro
37.
by sackfield 2y ago
Having optimism about the future means it will change for the better. You are referring to the present. Weird how many people find this concept uncomfortable, and perhaps is a good example of my original argument.
38.
by sackfield 2y ago
I've always thought the purpose of code review was to determine if the implementation was consistent from an architectural perspective, bug finding seems like something that should be caught by unit tests and other tools (unless you ha
39.
by sackfield 2y ago
In Dubai right now, one of the things I really like about this place is the optimism they have for the future. It's exhausting working in tech and being in places that don't believe engineering will usher in a better future (but s
40.
by sackfield 3y ago
I feel you, however I would say at the very least I find the YC submission process a very useful exercise and encourage all people thinking of doing a startup to complete it, even if you don't submit it. When I submitted to YC and got
41.
by sackfield 3y ago
The best content is not created by luddites. Creative curiousity is generally a necessity for great art.
42.
by sackfield 3y ago
It seems quite appropriate that luddites will opt out of the AI training space. Hopefully it will lead to better quality output.
43.
Meta to Wind Down Its News Feature in the US and Australia(bloomberg.com)
59 points by sackfield 3y ago|66 comments
44.
by sackfield 3y ago
It's refreshing to see an organisation act with restraint. It's easy to overstep authority by acting like a benevolent dictator, especially when people try and compress a complex topic down to "good vs bad", so it's
45.
by sackfield 3y ago
UBI experiments are mostly pointless until they reach a national level. Otherwise you end up giving people a bunch of extra cash from the working economy, and surprise surprise, people are generally happier and better off with more free cas
46.
by sackfield 3y ago
You are definitely right, this is a risky business: https://www.aljazeera.com/news/2023/4/21/qa-chilling-journal... Still, I wonder if there is a structure which resolves liability.
47.
by sackfield 3y ago
I wonder if a better model for journalism is the model employed by Hindenburg Research. They find stories that when published will have a devastating impact on a company, short the company, then drop their research and reap the change in ma
48.
by sackfield 3y ago
Thank god for index funds. Having said that a disaster is one classification and rebalance away.
49.
by sackfield 3y ago
They don't need to grow, efficiency can grow though as very aptly demonstrated in the 20th century.
50.
by sackfield 3y ago
I think the market has shown that growth is the solution rather than rationing. Lots of the humanitarian improvements of the 20th century (such as number of people starving) are improved due to market forces innovating on doing more with le
51.
by sackfield 3y ago
It's weird that people are acting like commodities are under some sort of license when they are purchased. As far as I know they aren't. A world that tries to regulate this scares me more than one where people participate in futil
52.
by sackfield 3y ago
I have noticed that there are a group of people actively pushing for these outcomes, to them this wasn't a mistake but a big victory. When they inevitably get criticised its very convenient to pretend like this was ordained by the gene
53.
Bazel: Aspirations and Realities(sackfield.substack.com)
3 points by sackfield 3y ago|1 comments
54.
by sackfield 3y ago
Some struggles I've had trying to implement Bazel in greenfield startups.
55.
by sackfield 3y ago
I have really tried to like Swift, and with new projects I do tend to use it, but legacy projects in Objective-C still make me smile that I get to go back to the old familiar language. I also like its interoperability with C++ through Objec
56.
by sackfield 3y ago
TPUs: https://cloud.google.com/tpu?hl=en
57.
by sackfield 3y ago
It's nice to know this also applies to the UK even after Brexit (still a member of the ECHR).
58.
by sackfield 3y ago
"At FoundationDB, once we hit the point of having ~zero bugs and confidence that any new ones would be found immediately, we entered into this blessed condition and we flew. Programming in this state is like living life surrounded by a
59.
by sackfield 3y ago
Is there a real market for this kind of LLM? Or is this the result of in-group preferences? EDIT: It is very very funny to talk to though.
60.
by sackfield 3y ago
If this is economically unproductive work, why are the miners being rewarded for it? Is it just that the market hasn't figured out what the detractors here seem to know? Or is it that they don't understand as much about economics

More ›