klysm

38 ms·

931.
by klysm 3y ago
You’re getting at just one axis of a larger trade off space. Yes serving scripts move compute cost to the browser, but you can also gain a lot of latency wins (which are sometimes critical) by having more compute on the client.
932.
by klysm 3y ago
I don't understand why all the arguments around frontend stacks on the web are treated with such dogma - it's a tradeoff space! You can send JSON, HTML, or whatever, but it depends on _what you are building_. The decision about wh
933.
by klysm 3y ago
Is spanner really practical for most people? From my narrow understanding it relies on some specialized clocks and networking
934.
by klysm 3y ago
A toast to all the time this year I have spent thinking through date time programming and lamenting calendars
935.
by klysm 3y ago
Strongly agree. Denormalizing data can sometimes be a good idea, but I don't consider this to be a reasonable justification.
936.
by klysm 3y ago
Cool! I feel like these could have nifty signal processing applications for hard real time systems.
937.
by klysm 3y ago
Is there an analog of an FPGA for analog computers? Or is that like what an analog computer is
938.
by klysm 3y ago
Obscurity can add some cost but has no real properties of strength.
939.
by klysm 3y ago
I mean it’s not stated as a strong claim, but a rough guess given the “perhaps”. I do agree saying an exact number with 0 basis is not a great argument though.
940.
by klysm 3y ago
It’s hard to compare given the intermittency, but I do agree the cost is a bit absurd. It’s painful because it doesn’t have to be this way but we’ve cornered ourselves via regulation into bad designs
941.
by klysm 3y ago
Has anybody made this for factorio
942.
by klysm 3y ago
The point of the bloat is that web is by far the easiest way to create a cross platform UI. It’s far from ideal but that’s the world we live in.
943.
by klysm 3y ago
Not sure the imposter syndrome is really relevant here. Hackathon is effectively throw a bunch of hackers in a room for a weekend or something and see what you can hack together
944.
by klysm 3y ago
Most systems are practically broken and it’s worked around by human factors
945.
by klysm 3y ago
But you wouldn’t need to lock them if repeatable read actually worked
946.
by klysm 3y ago
Completely agree, I've made that exact same argument on HN before.
947.
by klysm 3y ago
In my experience, most developers don't even consider isolation level in the first place and just take whatever the default is. Any race conditions are met with an 'oh that's weird', and then they move on.
948.
by klysm 3y ago
but CD
949.
by klysm 3y ago
I believe the idiomatic term would be “off topic” :)
950.
by klysm 3y ago
To elaborate, I want automatic strong typing for the query results
951.
by klysm 3y ago
I don’t know any ORMs except Zapatos which does what I’m describing with lateral nested joins
952.
by klysm 3y ago
You can do nested aggregates though unless I’m misunderstanding what you mean by nested aggregate
953.
by klysm 3y ago
I think this solves the problem of Cartesian explosions in ORMs. The requirement of related data needing to be shoved into a single table means data shared by many entities has to be duplicated. The result set being split into distinct tabl
954.
by klysm 3y ago
I love questions like this that feel like an XKCD
955.
by klysm 3y ago
Easier is exactly what that wouldn’t be
956.
by klysm 3y ago
Yes, but afaik CAN is not usually done over Ethernet
957.
by klysm 3y ago
I’m still skeptical. Why can’t it be filtered out
958.
by klysm 3y ago
I don’t understand how to calculate those odds though. Like is it a continuous rate of event probability?
959.
by klysm 3y ago
Why is it inherently noisier?
960.
by klysm 3y ago
Wouldn’t it be PoC instead of PoE?

More ›