Re:Launch HN: Dashblock (YC S19) – Turn Any Website into an API
by HPouillot on Wednesday September 18, 2019 @04:00PM
(#21009426)
It doesn't require Artificial General Intelligence, with enough training data (crowdsourced data and human-edited metadata like JSON-LD or RFD), we can classify automatically the attributes on the page (product name, movie title, creation date, author), structure them and recognise the type of entity.
Feel free to contact us if you want to invest ([email protected]), we are currently raising funds ;)
Re:Launch HN: Dashblock (YC S19) – Turn Any Website into an API
by rvnx on Wednesday September 18, 2019 @04:12PM
(#21009566)
But, what's the value compared to using open-source products like Portia https://portia.readthedocs.io/en/latest/getting-started.html ?
Functionally it looks very similar.
Re:Launch HN: Dashblock (YC S19) – Turn Any Website into an API
by r0rshrk on Thursday September 19, 2019 @11:36AM
(#21017189)
I'm sure this comment will go down in history like the Dropbox comment
Re:Launch HN: Dashblock (YC S19) – Turn Any Website into an API
by rvnx on Saturday September 21, 2019 @04:38AM
(#21034199)
Fine, I prefer to loose my comment than my invested money
Re:Launch HN: Dashblock (YC S19) – Turn Any Website into an API
by jakubbalada on Wednesday September 18, 2019 @03:58PM
(#21009404)
> A growing catalog of up-to-date scrapers for popular websites would put of lot of freelancers out of work. I would invest in this.
Check out Apify store (https://apify.com/store). It's built exactly for that purpose.
(Disclaimer: I'm a co-founder of Apify)
Re:Launch HN: Dashblock (YC S19) – Turn Any Website into an API
by enos_feedler on Friday September 20, 2019 @01:17PM
(#21029063)
Duplex for web [1] would certainly benefit from this kind of understanding so I wouldn't be surprised if Google is working on this.
1. https://www.theverge.com/2019/5/7/18531195/google-duplex-web...
However, the ML claim is highly suspect. There is no way that a machine could reliably understand the semantic content of a website - that would require Artificial General Intelligence. If anyone could do that, it would've been Google. But even Google relies on human-edited structured metadata to define the content of sites (i.e. Rich Snippets and the like).