Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
crorella
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
8 ms
·
61.
▲
by
crorella
3y ago
another nice option is Qobuz, the sound quality is good for the newer releases.
62.
▲
by
crorella
3y ago
Good thing there are better browsers out there. As long as they don’t mess with the transport we should be good
63.
▲
by
crorella
3y ago
Shorts have just downsides in my mind: - not enough time to provide quality content - dilution of engagement, because the lack of meaningful content you can produce and consume in a few seconds/minutes - further promote short span focu
64.
▲
by
crorella
4y ago
Worsening products to improve stakeholder value. But they forgot that the value itself is generated by good products that people actually want to use. Very shortsighted.
65.
▲
by
crorella
4y ago
Thanks!
66.
▲
Ask HN: Best resources to learn more about GPT, Chinchilla, LLaMa?
11 points
by
crorella
4y ago
|
3 comments
67.
▲
by
crorella
4y ago
What would you recommend instead?
68.
▲
by
crorella
4y ago
It seems there are portions of the content which is just a bunch of symbols, at least that's what the results are showing (ie. "±ÆÆßÚ ÔÏ œ ‹± ß±´ Wø™ª ø7ß ÆªøT±7 ¨± 檥1ª™ª ¨Wø¨ ø7ß ÔÎ ±J ß±´Æ \Æ1±Æ ") If I click the associ
69.
▲
by
crorella
4y ago
each team or project normally has a group where you can ask questions, provide feedback or ask for features.
70.
▲
by
crorella
4y ago
This is very interesting, as a Data Eng, I'd love to see what they are doing with the huge volume of data they collect.
71.
▲
Galactica: an AI trained on humanity's scientific knowledge (by Meta)
(galactica.org)
409 points
by
crorella
4y ago
|
358 comments
72.
▲
by
crorella
4y ago
I haven't checked recently, but last time I visited the site there were some ppl that is consistently posting high quality content, but it is like 5-10%
73.
▲
by
crorella
4y ago
If you unfollow everyone you don't see ads ever again. You just have to visit your friend's profiles one by one (or whatever group of friends you have created) to see their unaltered feeds with no ads.
74.
▲
by
crorella
4y ago
Relevant in the Trump case, as they are being charged for the same thing and they were complaining of political persecution since they were the only one being charged for that. Well, not anymore. EDIT: Context - https://www.wsj.c
75.
▲
by
crorella
4y ago
do you have a link or something with more context about them and what they are doing?
76.
▲
by
crorella
4y ago
hehe, I was thinking the same, then I read your comment. I wish they give support to Firefox in the future, getting text (and even modifying it!) is something I need to do often.
77.
▲
by
crorella
4y ago
it is back! I wonder what happened.
78.
▲
by
crorella
4y ago
What it means? It means people realized there is no inherent value in a tech the tries to solve already solved problems in am extremely inefficient way.
79.
▲
by
crorella
4y ago
What else to expect from a company whose entire business rests around a ponzi scheme? The way of backing up to people who already left their jobs was very unprofessional.
80.
▲
by
crorella
4y ago
I find lots of commonalities with other systems aimed at processing big datasets. The only thing I find different is their management of the metadata at a centralized place, which can bring lots of network IO savings since you don't ha
81.
▲
by
crorella
4y ago
I think overly confrontational people gets isolated sooner or later. There are certain times and places to talk more serious or pungent themes… like parties or sharing a beers or whatever, but not every single time you meet.
82.
▲
by
crorella
4y ago
I am extremely skeptical about this, I think college is better at that. In 17 years in the workforce I've made 0 friends, while I have 7-9 very good friends from college.
83.
▲
by
crorella
4y ago
> I wonder if you could get most of the way there by exposing the workload (across different pipeline stages) to a materialized view recommender yes! that's something we are trying to do, since we have a way to create signatures for
84.
▲
by
crorella
4y ago
Interesting facts about the tool, we created the parser with speed and efficiency in mind, it is able to process ~5M queries in less than an hour, since many of the SQL statements are associated with scheduled pipelines, I came up with a wa
85.
▲
by
crorella
4y ago
Yes! you can then put a UI on top of that and answer stuff like "who will get affected by my change on this table?" or.. "there is a bug in my pipeline and now I need to notify downstream consumers"
86.
▲
by
crorella
4y ago
haha, we did just that in my company, but aimed at distributed tables (like HDFS) for their partition and bucketing schemata.
87.
▲
by
crorella
4y ago
We created a custom purpose parser to get some interesting facts about the SQL statements running in our warehouse, a couple of millions of queries per day. With that information, we created a tool that finds the best partition and bucketin
88.
▲
by
crorella
5y ago
slightly unrelated, but it seems tripp [1] is also a nice approach to meditation using VR 1. https://www.tripp.com/
89.
▲
by
crorella
5y ago
I work in a similar area, have some questions: - How do you deal with standardization across the different events you log? How do you handle standardized naming, representations (formatting) and types? - What about validations? do you valid
90.
▲
by
crorella
5y ago
probably adding the hash means hundreds of $$ in storage
More ›