Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
rs23296008n1
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
14 ms
·
91.
▲
by
rs23296008n1
7y ago
Most engineering probably just requires more actual engineering. Too many people with "engineer" in their title have had little to no exposure to what engineering is. Especially software where we think we've done engineering
92.
▲
by
rs23296008n1
7y ago
This seems good to know... right up until some people do the usual silliness and overdose themselves out of their meat suit. Especially those taking meds they don't fully understand. Should be an interesting year ahead of us.
93.
▲
by
rs23296008n1
7y ago
Just a little project to let me talk to my lights and a few gadgets around the place.
94.
▲
by
rs23296008n1
7y ago
Rust syntax is strange but understandable to my C/C++/nim/python/elixir experience. I successfully managed to port a few test projects to it several months ago. Didn't mind the experience at all except for a few nig
95.
▲
by
rs23296008n1
7y ago
For me, a mild addiction to learning nim/wren and porting code to them. D and Rust can wait a little longer. People forget that there's a big world out there. We're all doing our thing.
96.
▲
by
rs23296008n1
7y ago
"Might pay" means no actual guarantee they will turn into a customer. They are a risk you need to investigate. You'll need to dig into actually why open source is a factor for them and what it really means. Are they just want
97.
▲
by
rs23296008n1
7y ago
Sharp edged words. High performance code only exists when you've developed experience from building low performance code. I think a few explanatory notes and discussion of suitable alternative approaches is warranted but not a complete
98.
▲
by
rs23296008n1
7y ago
Especially as I've seen plenty of goats on their hind legs. Excellent climbers. This book is a great idea.
99.
▲
by
rs23296008n1
7y ago
The model I use for some for-profit things I do: People who pay are your customers. Everyone else is free advertising / marketing / product or service testing / feedback group. Note that "who pay" doesn't neces
100.
▲
by
rs23296008n1
7y ago
Um, I'm talking writing the code on linux and targeting Windows using Linux cross compilation. The compilers are run on linux. The code is targeted at windows. You'd need windows to run it. Windows is the target .
101.
▲
by
rs23296008n1
7y ago
Cross compilation from linux targeting windows is possible. I'm only doing it for C++ but I'd assume others are doable.
102.
▲
by
rs23296008n1
7y ago
... and now I learn that NZ has done similar things to AU.
103.
▲
by
rs23296008n1
7y ago
Evidence. Replication of results by multiple unrelated investigators. Those two points alone remove a lot of gibberish. Even an appeal to authority is suspect if those first two aren't addressed.
104.
▲
by
rs23296008n1
7y ago
At least three factors regarding rationing of bandwidth: Data rate, data volume and caching. "Bandwidth" is already rationed but is usually high enough you don't notice most of the time or its still good enough most of t
105.
▲
by
rs23296008n1
7y ago
Having kids at home without an internet connection these days is something most parents would rather not experience. Right up there with lack of electricity but lower in the priority list. I'd have thought Netflix in particular would u
106.
▲
by
rs23296008n1
7y ago
While I think both countries' ISPs like to compete for a yearly "silliest restriction and worst customer experience" award, the NBN in Australia is a government screwup mostly due to the actions of vested interests and vested
107.
▲
by
rs23296008n1
7y ago
Still, if github really only provided a one liner "account blocked due to terms of service violation" and not much more explanation then github itself also rightly needs to be criticised. That by itself is bad behavior and general
108.
▲
by
rs23296008n1
7y ago
I'm sensing there is a lot more to this story. Related note while we're banning people for word selection: Does github realise that "git" has a pejorative association? Its not actually a polite word. If you were to call
109.
▲
by
rs23296008n1
7y ago
Don't know about "corporate shilling" - you weren't covert at all. I did ask for options after all. The only gotcha is the requirements will overwhelm most basic setups fairly quickly. Not talking 20 lights here or a tv
110.
▲
by
rs23296008n1
7y ago
Sure. We've got a house with multiple buildings, including sheds, halls etc. Around 100 people need separate profiles, each should be able to set alarms, timers, reminders, etc. if they want a routine to create any of those or tell the
111.
▲
by
rs23296008n1
7y ago
I think I'll dig in first. Cheers.
112.
▲
by
rs23296008n1
7y ago
Sounds good to me. RasPi are solid performers to us. I'm assuming the XML would need to be updated as the dictionary changes. Sounds easy enough. The loading of languages might get fussy/impossible if I want multiple. A stretch go
113.
▲
by
rs23296008n1
7y ago
If you've got an amazon echo or equivalent, try this experiment: disconnect your internet. Now issue a voice command to it. Why is your light not switching on? Why is alexa not answering your questions? Now you know why I want "of
114.
▲
by
rs23296008n1
7y ago
I think I will bother, the use cases are valid and worthwhile to pursue for multiple reasons, many of which aren't even on my radar. Bonus: I can have my cake AND eat it too.
115.
▲
by
rs23296008n1
7y ago
Sound great, thanks
116.
▲
by
rs23296008n1
7y ago
Exactly how mine works. Each language uses a different trigger. Simple. Even then it can misinterpret so plenty of room for improvement. My quick POC is around 90% accurate for language detection based on the trigger word.
117.
▲
by
rs23296008n1
7y ago
I'd expected some training required for each separately so we've got a decent collection of voice examples. Some of us mix languages within a sentence but that's a bad habit anyway so unsupported. I don't see figuring ou
118.
▲
by
rs23296008n1
7y ago
Looks interesting. I've got a few GPUs I could use if the CPU is too much of a bottleneck. We've identified a dictionary of the types of commands and words we use and have a recording of all our amazon and other commands. Training
119.
▲
by
rs23296008n1
7y ago
Its beginning to take shape. Already got a bunch of good candidates for experimentation next week based on the answers so far.
120.
▲
by
rs23296008n1
7y ago
Pertinent quote from [3] regarding "more challenging": "You must also deploy custom NLP models, as the official ones will not be compatible. Use this setup only if you absolutely need custom Thingpedia interfaces, and cannot
More ›