Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
lcampbell
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
11 ms
·
91.
▲
by
lcampbell
14y ago
I'm pretty sure the core game logic is not on the servers, as widely reported -- I was playing from 9-12PM CST last night, and the servers were supposedly completely down from 10-12PM -- I was able to continue playing in my city, and wasn't
92.
▲
by
lcampbell
14y ago
Where's this data coming from? Bitcoin Charts has different data: http://bitcoincharts.com/charts/mtgoxUSD#rg1zigHourlyzczsg20...
93.
▲
by
lcampbell
14y ago
The wildcard fix is annoying when you have everything on SSL but don't want to handle a wildcard cert[1]. When someone typos https://foo.example.com I'd like the UX to be a browser's "could not connect to server" error, not "this site is
94.
▲
by
lcampbell
14y ago
Why not alert = (args...) -> $globalMessenger().post args... This allows $globalMessenger to be notified for upkeep whenever the call to .post is proxied.
95.
▲
by
lcampbell
14y ago
Thanks for the reply. I figured for a dataset of that size, the main bottleneck might be not indexing -- maybe RedShift stores row data in a higher-latency medium while keeping indexes in-memory. Curious, I checked the documentation[1] and
96.
▲
by
lcampbell
14y ago
I really don't understand what's going on here. * You're measuring request latency. What part of that (for RedShift) is due to the network? (EDIT: I re-read and saw you're using `SELECT 1` as a gauge for round-trip latency and subtracting i
97.
▲
by
lcampbell
14y ago
Agreed. I take it a bit further: I have a dual-boot Windows/FreeBSD box, and use VirtualBox to run one OS inside the other, depending on what I'm doing (using a raw VMDK[1]). For me, this is helpful because using Windows as the primary give
98.
▲
by
lcampbell
14y ago
If you're talking about the in-client store UI, I believe it's PHP-generated markup rendered in an embedded WebKit browser. Given that the overlay browser uses the same tooling and also renders pages slowly, I'm inclined to believe they're
99.
▲
by
lcampbell
14y ago
No, it's just to explicitly inform the customer that Barracuda has shell access to the appliance. These accounts were undocumented.
100.
▲
by
lcampbell
14y ago
It looks and reads like a parody of filepicker.io.
101.
▲
by
lcampbell
14y ago
A quick google suggests that the patent sale price can be set arbitrarily, and the proceeds of the sale are simply taxed as either normal income or capital gains, depending on how the IRS feels that day. This corresponds with my thought tha
102.
▲
by
lcampbell
14y ago
There's a bit of detail in the wiki[1], you can probably get a better idea by searching the lists (there's likely a bunch on freebsd-current). Normally when someone brings up git on the lists, the common response is simply, "there's already
103.
▲
by
lcampbell
14y ago
> You can't wrap Lightning itself What prevents wrapping lightning? (I'm not at all familiar with Lightning) > any possible gains you could expect are lost at the CGO transition I was under the impression that the CGO overhead was d
104.
▲
by
lcampbell
14y ago
Those could be hand-entered default passwords for accounts that were never used.
105.
▲
by
lcampbell
14y ago
> DHCP assigned them the same IP, and google just continued the session which had been left open Given that you're using DHCP-assigned IP addresses, you're likely behind NAT. From Google's vantage point, both employees will appear to al
106.
▲
by
lcampbell
14y ago
I looked over the code because there was talk of switching over to OpenERP. My thoughts on the server component so far (keeping in mind that I'm exceptionally pedantic) -- - Internal inconsistencies w.r.t formatting -- sometimes there's new