Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
mNovak
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
28 ms
·
361.
▲
by
mNovak
5y ago
Not sure I'd suggest eastern Ohio, that's rural Appalachia, I would not recommend for interesting things.
362.
▲
by
mNovak
5y ago
Fun fact, Ohio was supposedly high on the Soviet nuke target list, because of Wright Patterson AFB, which houses the Air Force Research Lab (and aliens). I have an old Soviet invasion map of central Ohio, marking all potential airfields etc
363.
▲
by
mNovak
5y ago
Not sure if it's new, but I just recently filed a 5500 online. You can do it here: https://www.efast.dol.gov/welcome.html But yes, dealing with brokerage forms that needed a wet signature faxed..
364.
▲
by
mNovak
5y ago
Depending on where they're moving from, they might take a pay cut.. Ohio is very cheap to live in. New Albany is not particularly exciting, but nearby Columbus is a nice vibrant city, not the end of the earth. OSU is a definitely good
365.
▲
by
mNovak
5y ago
Lots of long term debt, like any other capital project. The $100B number is purely speculative, if the project grows over 50 years; they can build the first functional fab for $10B.
366.
▲
by
mNovak
5y ago
If you want to explore deeper, there's some fun (and well commented) open source GPS processing codes in Octave/Matlab [1], coming from the excellent book [2]. Those are compatible with cheap RTL-SDRs. [3] also looks nice, but as
367.
▲
by
mNovak
5y ago
I believe Wyoming actually passed some laws along these lines, to where a DAO can be a legal entity.
368.
▲
by
mNovak
5y ago
The author's description of 2015 doesn't sound like some ancient era to me.. Have things moved on and changed so dramatically from ML hype, Microsoft Edge, Apple Watch, and ES6? I'm not really here to argue about blockchains
369.
▲
by
mNovak
5y ago
Looks great, good luck! Bonus points for XKCD. I'm a big fan of technologies that bring new capabilities to everyone, not just specialists of that field. Minor clarification, I assume those 'includes X invocations' are per mo
370.
▲
by
mNovak
5y ago
Of these SaaS is still vastly the cheapest; even hardware costs aside, because the timescales involved are fundamentally shorter. Trying to raise capital for hard science (besides rockets and quantum apparently) is a real drag. We went to D
371.
▲
by
mNovak
5y ago
Another great site in this area is microwaves101.com A huge range of topics, and pleasantly snarky anonymous editor.
372.
▲
by
mNovak
5y ago
I assume you're referring to the Friis transmission formula? It only references frequency because that's usually more convenient than wavelength. I hardly think it's fair to call that scientifically illiterate. Also convertin
373.
▲
by
mNovak
5y ago
The wire transfer is just how the money exchanges hands. There would be a contract of some sort defining why that money is changing hands, and with what strings attached.
374.
▲
by
mNovak
5y ago
I will vouch that the Midwest is highly underrated. People are friendly; there's way more diversity and culture than people seem to think; it's affordable.
375.
▲
by
mNovak
5y ago
Just tried deleting my account--got exactly that error. That's not reassuring
376.
▲
by
mNovak
5y ago
The other half of what I'd consider "smart sync" is delta sync -- basically only uploading the bit of a file (or other reasonable small chunk) which has changed. Dropbox does this well, and is very useful if you're only
377.
▲
by
mNovak
5y ago
All I want is a dead easy way to make a landing page + account + payment options for a tiny prototype SaaS like this (e.g. I supply a few APIs as 'backend' and the rest just works) -- quite similar to the One Item Store the author
378.
▲
by
mNovak
5y ago
If crypto replaces Ticketmaster, that's a win for us all
379.
▲
by
mNovak
5y ago
Surprised actually this doesn't exist in a more general sense -- just X of N decryption of arbitrary files. Could be your private key recovery, or just mundane corporate documents with provable "two man rule"
380.
▲
by
mNovak
5y ago
The "proof" part of proof of coverage is very weak in Helium currently. Clearly most people are in good faith, but it is inherently vulnerable to Sybil attacks, because there's a chicken and egg problem with starting a new gr
381.
▲
by
mNovak
5y ago
Probably both sides can convince themselves the tech does or doesn't totally obviously beyond a doubt work (to differing definitions of "work"). But predicting public adoption and widespread use (which makes most things more
382.
▲
by
mNovak
5y ago
I never understood how a personal check could be secure. You can literally have them printed for you over the internet, and deposited via image. I just can't understand where the security comes from, it feels like a liability just havi
383.
▲
by
mNovak
5y ago
Actually very useful. It's sadly difficult to go from dxf (e.g. from a CAD software) to gerbers for PCB. Inkscape + svg2shenzhen handles it reasonably okay.
384.
▲
by
mNovak
5y ago
I think the arxiv / pre-print workaround could have so much more potential if more researchers were motivated/educated on it. I'd love to see an effort to simply email the top 10k authors to request preprint copies to be hos
385.
▲
by
mNovak
5y ago
Am I understanding this correctly; is the threat model that a block signer, some time later after liquidating their stake, can go and publish arbitrary versions of that older block?
386.
▲
by
mNovak
5y ago
Growing up in Ohio, this resonates with me
387.
▲
by
mNovak
5y ago
You're required to pay a "reasonable salary". So if you pay yourself minimum wage, but collect $100+k in profit, you may get funny looks from the IRS. Rule of thumb I've heard is 50% wages 50% profit, up until $100k wage
388.
▲
by
mNovak
5y ago
>> self-employed >> you got there through low stress My experiences differ
389.
▲
by
mNovak
5y ago
Someone already did this (create NFT of the permanent destruction of the original) with a Banksy original, so the art/shock value is spent. >> the most powerful act of preservation As if we don't already have scans / te
390.
▲
by
mNovak
5y ago
The loss on microwave relay is just incredibly high (think >90% power loss), because it goes as distance^2. Also there's physical limits on how tight the resulting beam of energy can be -- it's not a laser beam, it'd be ir
More ›