Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
nathants
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
12 ms
·
121.
▲
by
nathants
2y ago
this is awesome! i wish i had this before i knew aws well. workers+r2 is the ideal stack for 80% of problems now. to everyone complaining, just use workers+r2 directly. cheaper and more flexible. this is excellent ux around the much smaller
122.
▲
by
nathants
2y ago
in the us-west-2-lax-1a local zone, i just booted 100 r5.xlarge spot instances as fortnite like game servers[1]. 1 to be a central server, 99 to be fake players. the server broadcasts x100 write amplified data from every player to every pla
123.
▲
by
nathants
2y ago
this is cool! i do the same thing[1], wanted ag/rg to be able to search and replace. 1. https://github.com/nathants/agr
124.
▲
by
nathants
2y ago
assuming a serial writer, use a monotonic sequence that you can walk until no more data is found. for multiple writers, you have to use uuid and pay the search cost.
125.
▲
by
nathants
2y ago
you think reagent is more complex? i don’t even know what the difference between hooks and classes are, reagent has never given me a reason to care. if that’s not simplicity, i don’t know what is. if i were to build a 3d high performance ga
126.
▲
by
nathants
2y ago
that was when it rolled out to all regions. prior to that it existed in a single region only. not sure when that initially rolled out.
127.
▲
by
nathants
2y ago
in an immutable context, cas updates the content hash a named pointer points to.
128.
▲
by
nathants
2y ago
i do the same thing! i use spinning rust in x3 single usb external enclosures[1]. i mirror s3 data, both as an extra backup and to prevent s3 egress billing for random access. s3 egress is for disaster recovery only. 1. https://g
129.
▲
by
nathants
2y ago
the common approach is cas dynamodb pointers to uuid named objects in a named s3 prefix. would it be better to merge ddb and s3? maybe, maybe not. gcp/azure/etc are there to provide fancy cloud solutions. no need for aws to serve
130.
▲
by
nathants
2y ago
the correct pattern is to append immutable objects into a common prefix.
131.
▲
by
nathants
2y ago
s3 is not perfect, but close enough. the ideal design is to use s3 for immutable objects, then build a similar system on ec2 nvme for mutable/ephemeral data. the one i use by default is s4[1]. 1. https://github.com/nath
132.
▲
by
nathants
2y ago
no, i’m not interested in seo. this setup is a single go binary that serves a single pre-gzipped html file from a lambda. that html file contains inlined js, that is a reagent app. there is also a favicon. total of 3 files in the lambda zip
133.
▲
by
nathants
2y ago
the best reason to learn clojure is reagent[1], by far the best way to use react. shadow-cljs[2] makes using npm libraries easy. i’ve settled on go backends and reagent frontends as my default setup[3]. 1. https://reagent-project
134.
▲
by
nathants
2y ago
does it suck less though?
135.
▲
by
nathants
2y ago
imagine teaching kids to play this game instead of teaching them to self study with gpt4, kagi, and all the other tools revolutionizing information access. when your brain is tired, go outside and make your body tired too. with all the time
136.
▲
by
nathants
2y ago
isn’t this what —ff-only is for?
137.
▲
by
nathants
2y ago
postmarketos is a good place to start.
138.
▲
by
nathants
2y ago
run mighy-snitch[1] and you’ll be a little more secure. aside from mandatory access control, for which there isn’t a usable solution for the typical desktop setup, monitoring network access attempts is as good as it gets. you’ll be surprise
139.
▲
by
nathants
2y ago
never not using littlesnitch. everything is always phoning home, even innocuous things like firefox.
140.
▲
Show HN: Immutable disk mirror so simple that unborkable
(github.com)
1 points
by
nathants
2y ago
|
0 comments
141.
▲
by
nathants
3y ago
hetzner egress is x100 cheaper. it’s a good reason to switch.
142.
▲
by
nathants
3y ago
due to shareholder primacy, one can no longer seek investors. seek customers instead. this is fine.
143.
▲
I'm building a better Fortnite. ask me how [video]
(youtube.com)
1 points
by
nathants
3y ago
|
0 comments
144.
▲
by
nathants
3y ago
this is why you run hellzerg/optimizer on install and never update windows again. sigh…
145.
▲
by
nathants
3y ago
i print the error along with file and line number every time i return it. clunky, but it works. in fact i print file and line with every log message. https://github.com/nathants/libaws/blob/87fb45b4cae20abd1bb
146.
▲
by
nathants
3y ago
it’s not. stuff logs into s3. learn to mine them in parallel using lambda or ec2 spot. grow tech or teams as needed for scale. never egress data and never persist data outside of the cheapest s3 tiers. expire data on some sane schedule. dat
147.
▲
by
nathants
3y ago
the online game indie renaissance is coming. it’s just so much easier to build online games than it used to be. a lot of gamedevs and software engineers getting laid off these days. some small percentage of those is gonna go build things in
148.
▲
by
nathants
3y ago
thanks! it’s incredibly fun and challenging. would be hard to stop.
149.
▲
by
nathants
3y ago
i’m building a better fortnite, exploring new technical frontiers in online gaming. lower latency, higher bandwidth, better gameplay auditing, etc. it’s so fun to see network degradation or bugs as ingame visuals and instantly understand wh
150.
▲
by
nathants
3y ago
run littlesnitch or something similar to notice and prevent egress attempts. for now it seems the only effective defense. hopefully somebody builds a disk snitch. would love to buy that.
More ›