Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
gbrits
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
8 ms
·
31.
▲
by
gbrits
6y ago
A Geohash ( https://en.wikipedia.org/wiki/Geohash ), a (decoding of a) recursive spacing-filling curve, would be much faster still: you literally slash a couple of characters of the tail of the hash to end up with your b
32.
▲
by
gbrits
6y ago
Currently using Minio which is a relative success. However as part of quality checks we're using a lot 'listObjectsV2' S3-calls to check for gaps in our data, which is pretty slow. When using SeadweedFS plus, say, a Filer met
33.
▲
by
gbrits
6y ago
Comment as bookmark
34.
▲
by
gbrits
6y ago
It's the discount. E.g.: $10 buys you a $12 gift card .. buys you a $14.4 gift card... ad infinitum
35.
▲
by
gbrits
7y ago
Facinating. There's a lot of Dutch people here. Some might want to translate
36.
▲
by
gbrits
7y ago
“and that IRR includes crashes like these.” No it doesn’t.
37.
▲
by
gbrits
7y ago
Agreed. It really blows generic compression algos out of the water. Didn't know about Zstandard dictionary encoding. This might just be what I'm after. Thanks
38.
▲
by
gbrits
7y ago
Slightly related: recently wrote a binary encoder/decoder that used bitpacking, delta-encoding, and other well-known 'tricks' to efficiently pack batches of 10k-100k events of the same uniform type (essentially encoding colum
39.
▲
by
gbrits
7y ago
How so? Haven't tried tailwind, but wanted to try it on a pretty decent-sized Vue-project
40.
▲
by
gbrits
7y ago
Looks promising! Can materialized views be backfilled?
41.
▲
by
gbrits
7y ago
Is this similar to TimescaleDB's Continuous Aggregates? Interested in knowing the overlap / differences.
42.
▲
by
gbrits
7y ago
Do any of these autoML offerings have a way to use the generated model in JavaScript/nodejs? I know of [sklearn-porter]( https://github.com/nok/sklearn-porter ) which transpiles scikit-learn models to JavaScript amo
43.
▲
by
gbrits
7y ago
I follow everything else but what's the health benefit of avoiding meat?
44.
▲
by
gbrits
8y ago
So the market maker who pays for order flow data takes the spread. But how are they able to trade with the buyer at $100.01 when there are limit asks on the book already at that price? Seems they need to cut the line?
45.
▲
by
gbrits
8y ago
To solve connect to a connection pool service instead of a RDBMS directly. E.g. Pgbouncer for Postgres
46.
▲
by
gbrits
9y ago
A bit at a tangent, but I was recently looking for software (open source or commercial) that could help me develop a 'interior design / planning' editor. I.e.: given a specific area (outside dimensions), allow to drag walls,
47.
▲
by
gbrits
9y ago
emphasis on should
48.
▲
by
gbrits
9y ago
So where's the point & click GUI to select items from a page? Signed up, but can't seem to find it on first glance
49.
▲
by
gbrits
9y ago
Assuming you want this done automatically, what's the advantage of 'printing' multiple tabs to PDF in a headless browser, over just sequentially loading and printing the pages you want done?
50.
▲
by
gbrits
9y ago
Using managed queues (SQS) from AWS. The rest is on Vultr. That mix-and-match works perfectly for my use-case. In other words, no need to be able to run everything on a single platform to get cost gains.
51.
▲
by
gbrits
9y ago
Agreed, although Mapbox comes close imho.
52.
▲
Ask HN: Why isn't there a Saas-for-Saas?
10 points
by
gbrits
10y ago
|
3 comments
53.
▲
by
gbrits
10y ago
Could this be used for record deduplication? Say you have products from various data sources. A product has various features/facts that almost always don't line up 100%. Could this library be used to match over such records? One o
54.
▲
by
gbrits
10y ago
This is solid: http://www.informatik.uni-hamburg.de/TGI/PetriNets/
55.
▲
by
gbrits
10y ago
Interested to know where to get 100k proxies for $100/mo. Can you give some options?
56.
▲
by
gbrits
10y ago
How about using the new AWS Step Functions instead? This would allow a full push model while still keeping a nice and tight workflow: 1. fire off AWS Step function on new post to S3 bucket 2. read delay / schedule time in metadata fro
57.
▲
by
gbrits
10y ago
Network latency would be a valid concern for avg. time/req. not for throughput (necessarily)
58.
▲
by
gbrits
10y ago
Thanks. Reading the 'problems' section: a simple solution for "Resilience against crashing or hanging workers" would be having a TTL on the lock. If only...
59.
▲
by
gbrits
10y ago
was it unusable during resize?
60.
▲
by
gbrits
10y ago
> I am actually planning on taking what I have learned on this and releasing it soon. Consider me interested. Any way I could be kept in the loop?
More ›