Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
paragraft
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
11 ms
·
61.
▲
by
paragraft
8y ago
Their big reinvent conference is this week, they're making a series of product announcements.
62.
▲
by
paragraft
8y ago
GP used the wrong term, think they meant adaptive capacity, which is a newer feature where shards will automatically lend capacity to each other in the case of hotspots.
63.
▲
by
paragraft
8y ago
Planet Money had a great two part episode[1] on this and how it came about, and the guru who advocated it. Particularly fascinating was the gigantic PowerPoint presentation they embedded that formed the pitch deck for the proposal. It'
64.
▲
by
paragraft
8y ago
I don't think it is, but—as usual with EC—if you don't want it there's a performance cost. The way you'd avoid it in a model where writes go into one queue/connection and reads come out somewhere else is when you&#x
65.
▲
by
paragraft
8y ago
Conflict resolution is a separate problem that isn't addressed by Eventual Consistency. If I make a write in an EC system, that write may eventually be accepted, it may be rejected, or it could be resolved through something smarter lik
66.
▲
by
paragraft
8y ago
Starlink will only be most useful in remote areas, as you'll be sharing the satellites above you with fewer people. In municipal areas terrestrial will always win.
67.
▲
by
paragraft
8y ago
No they charge per platform. I've bought it 3 times (Android, iOS, web).
68.
▲
by
paragraft
9y ago
By "one way" the OP was referring to the ground-to-satellite trip taking 110ms. So by "round way trip" they meant a packet going from client to satellite to ground station (2 x 110ms). Thus 440ms.
69.
▲
by
paragraft
9y ago
Guardian is wrong to use "overshoot" as though it was a mistake - it was a burn to exhaustion, the exact limits of which was unknown beforehand (because it was unknown how well the fuel in the second stage would survive the time i
70.
▲
by
paragraft
9y ago
Elon said in the press conference afterward it was just a mannequin in the suit, no instrumentation.
71.
▲
by
paragraft
9y ago
He means the floodlights (usually motion-triggered) attached to his own house.
72.
▲
by
paragraft
9y ago
You don’t remember the pre-FB era of fwd’ed email chains? They were Snopes’ original raison d'être.
73.
▲
by
paragraft
9y ago
There'll be a small chunk of the purchase price left in escrow for a year for any extra liabilities that weren't discovered in DD. They'll be claiming that. But it won't be much.
74.
▲
by
paragraft
10y ago
Those limitations apply to the old model of App Engine they call the Standard Environment, which abstracted much more over the underlying resources (you didn't have to think about what VMs you were using). At some point they realised t
75.
▲
by
paragraft
10y ago
Kiln from Fogcreek. Slow on a good day, it just mysteriously stopped noticing my commits in my activity stream some four years ago, you can't have reviews spanning multiple repos, they've blown off pull-requests as a feature for y
76.
▲
by
paragraft
10y ago
I'm in no position to discuss the actual service, but was really surprised by the citation requests at the bottom. If you use the service, please cite these 6 papers. If you use the data, please cite these other 2. Is this a new norm t
77.
▲
by
paragraft
10y ago
What a great post. I really appreciate these longer digs into the past that go behind the 'what' to explain the how & why of where we got where we are; and just as much the futures that could have been and why they didn't
78.
▲
by
paragraft
10y ago
Neat idea. Do you know why it missed Nantucket? (Gets heavy mention in Moby Dick).
79.
▲
by
paragraft
11y ago
Consider that those fictional characters have professional writers doing all their lines. It helps a lot.
80.
▲
by
paragraft
12y ago
As well as the one click limit, there's an easy fix for this assuming your kids aren't aberrantly tall for their age.
81.
▲
by
paragraft
12y ago
Will prefix this with the necessary disclaimer that there are exceptions to this on both sides, and by generalising I'm necessarily going to get it wrong for some specific cases. One of the best devs I know dropped out of CS in the fir
82.
▲
by
paragraft
12y ago
Do basic CS concepts pertain to the job? Well that depends on your job I guess. I consider them the tools of the trade that will outlast all the languages and frameworks that will come and go in a career. And you can intuitively learn them
83.
▲
by
paragraft
12y ago
Had a great opportunity to use this in anger a few months ago. Working on a mapping application, where the map could only handle a few hundred items at a time, but the customer account had tens or hundreds of thousands. And the user could s
84.
▲
by
paragraft
12y ago
I read an excellent biography/history on Kim Philby ( http://www.amazon.com/Spy-Among-Friends-Philby-Betrayal-eboo... ) where the author drew on this essay in an attempt to explain Philby's motives for treason. Born
85.
▲
by
paragraft
12y ago
At a guess, scrub fire hazard?
86.
▲
by
paragraft
13y ago
There's a product in the Australia and NZ market called One Square Meal which seems exactly this (which would seem to raise the spectre of trademark conflict). http://www.onesquaremeal.com/productinfo.html For somethin
87.
▲
by
paragraft
13y ago
I'm not sure about on the Mac, but that sounds like the MS Works 1.0 tutorial for the database component. The scenario was for a ski field where there had been a murder, and you had to whittle down the suspects based on the clues found
88.
▲
by
paragraft
13y ago
Its list handling seems broken? tojson(fromjson(x)) !== x. Paste { "name": "John", "age": "29", "foo": [1,2,3] } into the JSON side, and click in the left and move the cursor to force it to
89.
▲
by
paragraft
13y ago
Why is the countdown stuck at T-00:13:00? Shouldn't it be counting down? Edit: Nvm, saw the overlay saying they're on a hold.
90.
▲
by
paragraft
13y ago
Reminds me of my favourite Javascript trivia question: What two values for a & b meet the following? > a === b true > Number(a) === a true > Number(b) === b true > a + b === a true > (1 / a) === (1 / b) false
More ›