estebarb

20 ms·

241.
by estebarb 4y ago
Oh is ridiculous because Twitter does nothing with accounts that only exist to perform harassment or hate speach.
242.
by estebarb 4y ago
At our university we do the clarification ourselves, usually with 3-5 classifiers per item. And it is surprising the high rate of disconformity between labelers (even in binary or 4-class classification). People in internet don't alway
243.
by estebarb 4y ago
I find it ridiculous because nobody is forced to use it or tag people so the model can find others. And afaik it only works for personal usage. Is like getting Apple/Google demanded because they allow to dedup contacts.
244.
by estebarb 4y ago
Nice, it looks a little bit like Micromundos (they translate the logo language). Personally I would prefer a C like syntax instead of Algol.
245.
by estebarb 4y ago
It is dictionary compression: https://en.m.wikibooks.org/wiki/Data_Compression/Dictionary_...
246.
by estebarb 4y ago
Humans are pretty good understanding texts with missing letters. I guess that if it considers also the context it would be possible to predict better the destroyed words. But basically, we would be storing the missing information in the dec
247.
by estebarb 4y ago
It is impossible to create a perfect hash function, unless you have the complete dictionary (otherwise there is collision risk). If you have the full dictionary then there are perfect hash algorithms. So, in essence, you want to add metadat
248.
by estebarb 4y ago
The date is when the url was first crawled, not the date of the current content.
249.
by estebarb 4y ago
There is a lot of public literature on how to build a search engine: it is not that secret. You must crawl millions of websites, then process them, index, deduplicate, remove spam... And then you must create a frontend client that queries t
250.
by estebarb 5y ago
Technical meetings are going to be fun with hispanic devs... "And finally we sent the metrics to Mimir /giggles/" Sadly they don't support encryption at rest (sorry, I really had to do one more pun)
251.
by estebarb 5y ago
> The data is then encrypted before it can be used to train AI models How can they train ML models with encrypted data?
252.
by estebarb 5y ago
As a programmer I think that this is awesome. But... I'm also thinking that we may be missing opportunities. For example, I prefer to see subtitled movies in their original language (I'm a Spanish native speaker). But there are so
253.
by estebarb 5y ago
For common programs it's not very difficult, usually just recompile. But if it uses assembly then you are out of luck. Also if it requires a library that uses it. For an OS it is different, you need to reimplement many parts: bootloade
254.
by estebarb 5y ago
Some funny example is "Eclipse": Do you mean Eclipse the IDE? Or the vampire movies? Maybe you mean the astronomical event? or the Nissan Eclipse car? Or you wrote wrong ellipse? I'm fine with Google not reading my mind at th
255.
by estebarb 5y ago
The monthly snapshots of CommonCrawl contains around 64k files of HTML, metadata and text only. The text only and metadata are around 250MB each, so in total the corpus would be around 16TB of text and 16TB of metadata. I don't know ho
256.
by estebarb 5y ago
In Spanish people gets two days bans by asking for "tinta HP negra" (black HP ink. In Spanish, HP not only refers to HP Inc, but shortcut of 'son of a bitch'). But if you run a fake news page promoting a coup d'etat
257.
by estebarb 5y ago
I worked on a database engine that uses JSON as communication protocol. Sure, most messages are small, but for a system processing JSONs 24/7 it quickly sums a lot. The main issues see had with our codec was memory allocation and the b
258.
by estebarb 5y ago
I'm hoping that the next stop is Lambda on GPUs
259.
by estebarb 5y ago
There are languages like Go, Rust or C that are compiled. If your lambda uses compiled languages you need to recompile to the right platform. Additionally, a lot of interpreted languages depends on native libraries. You need to build those
260.
by estebarb 5y ago
It doesn't make sense at all. When I'm driving a car, let's say a Toyota, I expect that the car would protect me from its own failures (like: alerting me of oil or battery issues). But no way I would expect the car to avoid c
261.
by estebarb 5y ago
Let's sue the USA? It's name implies that the states are united, but Hawaii and Alaska are not adjacent to the other states.
262.
by estebarb 5y ago
That is obvious sarcasm. I really think that Twitter is trusting too much on NLP for automating those checks and not humans. Because machines are particularly bad at grasping sarcasm. Its funny, because if you report obvious self harm or vi
263.
by estebarb 5y ago
Yes, because in other languages things costs less body parts. In Spanish, for example things just costs an eye or a kidney. Spanish -> literal translation Me costó un ojo -> it costs me an eye Me costó un riñón -> it costs me a kid
264.
by estebarb 5y ago
The joke is completely understandable to anybody that speaks Spanish too. In fact, many languages have idioms about exchanging body parts for something.
265.
by estebarb 5y ago
Mostly that I started my crawler before learning about Colly and it didn't make sense to rewrite the code. By "not Colly" I just wanted to remark that in Go is relatively easy to write a crawler from scratch.
266.
by estebarb 5y ago
"A wheel? Don't be silly, how are you going to add wheels to a horse?" (Prehistoric HN commenter)
267.
by estebarb 5y ago
It is very slow. But personally, I prefer to write my crawlers in Go (custom code, not Colly).
268.
by estebarb 5y ago
Materialize ( https://materialize.com/ ) is capable of performing SQL operations over a stream, using incremental calculation based on differential dataflow. CockroachDB is a distributed SQL database, with strong consistency.
269.
by estebarb 5y ago
Given that this kind of stories are each time more frequent: why are we still buying Apple? Its not like the hardware is super special or the Unix compatibility (that we already have in Linux or even Windows).
270.
by estebarb 5y ago
This is extremely cool!!! Also I find surprising that, as a Spanish speaker with no whistle-spanish training, I was able to understand most of the whistles.

More ›