Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
cr0sh
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
65 ms
·
481.
▲
by
cr0sh
7y ago
While you can still do what you did, likely you wouldn't have a static IP - at least not here in the United States. Most home connections have a dynamic IP provided by DHCP; now, usually, that IP won't change (probably tied to the
482.
▲
by
cr0sh
7y ago
I noticed that, too. I don't understand why Octave is this "dark horse" when it comes to these kinds of things...
483.
▲
by
cr0sh
7y ago
The term "Arduino" is for the ecosystem; it doesn't refer to any singular microcontroller. At one time, long ago, it did - mainly the ATMega8 platform that was at the heart of the Wiring project, which ultimately begat the or
484.
▲
by
cr0sh
7y ago
Surely someone by now has created a distributed video streaming system coupled with search? Maybe based on bittorrent or something? It seems like an obvious solution to this issue. It wouldn't be fast, and videos might not be as "
485.
▲
by
cr0sh
7y ago
After videos on repairing stuff gets outlawed - what's next? Videos on how to build stuff. Want to watch a video showing you how to build a quadcopter that doesn't have any kind of geofencing controls? Not allowed.
486.
▲
by
cr0sh
7y ago
Step 1: Obtain blue wrench. Step 2: Turn on blue wrench. Step 3: Apply blue wrench to padlock. Step 4: Profit. /can't be locked if it's a liquid...
487.
▲
by
cr0sh
7y ago
Sure - there may be nothing wrong with his IRS tax filings; he may have gains or losses listed under "other" - but what does that mean? He reported them, and that's all that matters, right? Except for the fact that there was
488.
▲
by
cr0sh
7y ago
> They need a warrant, the IRS can’t proactively share tax info with law enforcement agencies for non-tax law reasons. Unless the US Congress asks for them.
489.
▲
by
cr0sh
7y ago
A good way to experience the Amiga (and much cheaper than trying to get the relevant real hardware together) is to purchase and install a copy of the AmigaForever distro: https://www.amigaforever.com/ Technically, you could
490.
▲
by
cr0sh
7y ago
It's been mentioned in passing form a few times, but if the machine is decent enough (cpu and ram), you might consider setting it up to run and learn Docker or some other virtualization/container system.
491.
▲
by
cr0sh
7y ago
> especially if it had a relevant GPU For basic learning on somewhat larger datasets (sub-TB), I have found that my 750 TI SC GPU has been very adequate; the only time I maxed it out was while doing "live training" of a tensorf
492.
▲
by
cr0sh
7y ago
$30 a month should not be a hardship for most; I think I spend more for fast food every week (I certainly spend way more than that for random ebay and amazon purchases each month - and let's not get into what I spend on my Jeep). I can
493.
▲
by
cr0sh
7y ago
So backups are against the law in the UK? How does that work? Note: You can digitally "rip" an audio CD as a digital file; technically you are making a backup of the data that forms the audio on the CD. Provided you are only doing
494.
▲
by
cr0sh
7y ago
For a while I ran a box as a combo LAMP and ZoneMinder server; I ended up shutting it down as I wasn't using the LAMP server much any longer, and ZoneMinder was overkill for my purposes... ...but if you've never played around with
495.
▲
by
cr0sh
7y ago
> Even parallax based systems are good at avoiding stationary objects, and are inexpensive since they use two overlapping visible light cameras (see EyeSight for one example). Essentially they work like the human eye to determine depth&#
496.
▲
by
cr0sh
7y ago
> The main problem is not that we'll eventually run out of oil. The concept of "peak oil" (discounting whether or not it's a fallacy) wasn't ever that we'd run out of oil. When peak oil occurred, there would
497.
▲
by
cr0sh
7y ago
"Mind Matters is published by the Walter Bradley Center for Natural and Artificial Intelligence." The "Walter Bradley Center for Natural and Artificial Intelligence": https://centerforintelligence.org/ .
498.
▲
by
cr0sh
7y ago
I wonder if you looked at time estimates from developers of common process applications (like CRUD) in say, the 1980s, who used COBOL or FORTRAN, vs say those using Pascal or C - if those developers of code using extremely well known proces
499.
▲
by
cr0sh
7y ago
This would've worked if you'd taken your initial estimate (1 week) - bumped it to the next unit (1 month) - then doubled it (2 months). Maybe you would've been given 9 or 10 weeks, so you'd have the week for X, and 1 mon
500.
▲
by
cr0sh
7y ago
Has there been any work or thoughts on what to do for long distance city-to-city hops? Like say you wanted to connect the NY Mesh to Boston (not sure if they have one, just an example)? I live in Phoenix - I don't think we have anythin
501.
▲
by
cr0sh
7y ago
Most microcomputers from the 1980s were definitely as or more powerful, at least from a memory footprint and CPU capability standpoint. A PC keyboard controller - I guess it depends on what is being used; I suppose something that can read a
502.
▲
by
cr0sh
7y ago
I would publish it immediately, after checking as best I could that it was actually a working solution. But here's the twist: I would publish it in such a way that it would appear to have been released/solved by the worst person I
503.
▲
by
cr0sh
7y ago
I believe I am fairly "mindful" - but I am also a huge procrastinator, especially on tasks I don't really want to do. Maybe I'm not as mindful as I think? What's also interesting is that - for certain of these tasks
504.
▲
by
cr0sh
7y ago
This is a very open-ended question; depending on what you want to do with such an arm would be important to know what to recommend. Do you need fine control and accuracy? Do you need speed, or strength, or some combination? What kind and si
505.
▲
by
cr0sh
7y ago
So you're in disagreement with Hinton? https://www.axios.com/artificial-intelligence-pioneer-says-w... I've heard the argument made about "in memory compute" - and I do know that various companies have m
506.
▲
by
cr0sh
7y ago
Note: I am not an expert on ML, ANNs, etc - most of what I have done has been mainly "hobby level" and MOOC learning. From what I understand, though, what GH is trying to accomplish with capsule networks (which I have tried to und
507.
▲
by
cr0sh
7y ago
I'm biased, because I took the course when it was called "ML Class" in 2011: https://www.coursera.org/learn/machine-learning Everything is done in Octave (ie - open-source matlab-like language); primitiv
508.
▲
by
cr0sh
7y ago
I'm not a physicist, nor a philosopher, and certainly not a mathematician - so maybe what I'm going to write will seem both naive and probably wrong, but I'm doing it anyhow. In this Quanta interview, Amanda Gefter writes: &q
509.
▲
by
cr0sh
7y ago
She may outgrow it, but she'll probably never forget it. I'm a guy, but I recall as a kid feeling my dad's stubble when I gave him a hug as a child, when he came home from work, smelling of sweat and tar (he worked as a road
510.
▲
by
cr0sh
7y ago
When I see articles on these kinds of materials - handmade and fairly unique, but not in wide demand any longer - I tend to wonder: Are there any applications for the techniques used in the making of the materials that could be applied to o
More ›