Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
xondono
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
41 ms
·
451.
▲
by
xondono
5y ago
Not to be dismissive, but how does that differ from any other consulting/work for hire job? Most projects in embedded devices are exactly the same. - a price tag is fixed by negotiation around estimates - more and more stuff is pushed
452.
▲
by
xondono
5y ago
“We can’t make it work as a computer, so we made it look like a computer”
453.
▲
by
xondono
5y ago
That’s where the “hop around” comes in.
454.
▲
by
xondono
5y ago
A LOT of people here seem to be very willing to tax unrealized gains. I hope you all know that that would kick gentrification into hyperdrive and force a whole lot of people out of their homes.
455.
▲
by
xondono
5y ago
My team handles all code, documentation and release through Github. I was asked (more like forced to) have someone in my team copying zips of our releases and converting all code documentation into google doc files (with company logo, pro
456.
▲
by
xondono
5y ago
To me the most interesting part of this debate is that the majority of commenters here have pretty quickly jumped to “merit is luck because intelligence is luck”, which it’s pretty much impossible to make compatible with the concept of free
457.
▲
by
xondono
5y ago
As someone who suffered it, it was way worse. There was “ingeniería informatica”, which was kind of CS, but focused highly on software (databases were a significant part of the curriculum). Then there was “licenciatura informatica”, alt
458.
▲
by
xondono
5y ago
Using NN for cache management is not that new, and is a simpler approach in comparison to Intels as far as I know.
459.
▲
by
xondono
5y ago
I don’t know if it’s “cheap” but It’s probably cheaper than the monolithic alternative right?
460.
▲
by
xondono
5y ago
In short, yes, higher capacity will involve higher latency. The thing here is balancing core speed vs “effective” memory latency (factoring in cache hits, misses, etc). Cache management is a hard problem, and the equilibrium point is load d
461.
▲
by
xondono
5y ago
“Simpsons paradox is rare” might well be true, but is quite the stretch from what the author shows. Self reported opinion data is a very specific dataset to begin with.
462.
▲
by
xondono
5y ago
PWM periods and on-time are implemented in the time domain in hardware, so my guess would be issues with integer algebra. Some hardware implementations are notoriously weird on edge cases. Some manufacturers built PWM modules so 0 setting o
463.
▲
by
xondono
5y ago
1.5ºC is the lower bound for the estimated target. In any case, the 2ºC nominal target was mostly arbitrary. For those unaware, the problem with measuring global temperatures is that we can measure variation (relative values) with great acc
464.
▲
by
xondono
5y ago
Probably quite poor. FR4 boards are built with woven fibers, and you can get measurable differences depending on how tight the weave pattern is, or the angle of your transmission line relative to the weave pattern. Given that paper is not
465.
▲
by
xondono
5y ago
Maybe it’s me, but I think “fierce” has thrown a lot of people off. I would classify what you are describing as what PG terms “the bitter nerds”, not necessarily “fierce”.
466.
▲
by
xondono
5y ago
Nvidia is getting nice profits, and with way less risk than those “investing” in crypto
467.
▲
by
xondono
5y ago
> Netflix is already paying it's ISP And I would argue that that is good . It creates an incentive for Netflix to invest into encoding, compressing and developing new tech in general. It also provides openings for competing platfor
468.
▲
by
xondono
5y ago
Google and Netflix did campaigns for NN. So did Facebook and other internet and media companies. One of the most successful was the John Olivers Last Week Tonight piece. I saw some evidence later that the ‘evidence’ provided in that piece
469.
▲
by
xondono
5y ago
Yes there is, because none of what you mentioned is “the problem”, the problem is that all these practices are deliberately put into practice without users consent, or by trying to trick users into inadvertently give consent.
470.
▲
by
xondono
5y ago
Wasn’t this posted some time ago?
471.
▲
by
xondono
5y ago
Location: Barcelona Remote: Preferred Willing to relocate: Yes Technologies: Electronics and Embedded development, High Speed Design(DDR4, DisplayPort...), Experience with machine learning in embedded devices. Résumé/CV: h
472.
▲
by
xondono
5y ago
> Don't bother contacting me if you work in or for the car industry. Good to see I'm not the only one
473.
▲
by
xondono
5y ago
I’ve said this before, and even looked into it, but my abilities with web/app development are non existent. Can we get a app/service to read these in some sensible way? That would be awesome.
474.
▲
by
xondono
5y ago
This is nothing new, it’s just the MBA syndrome. Go to any MBA school and they’ll tell you how sales is a clear “profit center”, while engineering is a “cost center”, and that companies are made more profitable by “internalizing profit cent
475.
▲
by
xondono
5y ago
This. I feel like this is the manufacturing equivalent of the financial sector shenanigans back in 08. A bunch of companies increasing profits by baking in risks into the chain, because the risks of your individual providers failing are ind
476.
▲
by
xondono
5y ago
I’m just hoping we get soon to the collapse of bitcoin so we can either normalize better cryptos or just abandon the concept of cryptocurrency as a whole. Right now we are just inflating the proverbial balloon until it pops.
477.
▲
by
xondono
5y ago
Well, the primary difference is that cloud providers go through HDDs almost as consumables, while RAM and SSD are generally more long lasting. Contracts are set up in similar ways, but HDDs in general have more constraints baked in those co
478.
▲
by
xondono
5y ago
I’m not saying they shouldn’t. I’m just saying you should not be surprised of their complaints.
479.
▲
by
xondono
5y ago
This. Everyone is trying to milk the government for “support”, especially if they can get it in cash.
480.
▲
by
xondono
5y ago
Yes, that was implied in my comment
More ›