Hacker Newsnew | past | comments | ask | show | jobs | submit | more lawrencegs's commentslogin

I thought I am the only one


And here we are discussing this on HACKER NEWS :))))


Don't think I'll be adding this "social media" account information to any visa request.


then you'll be committing a felony


Not if you don’t consider it social media. Which I don’t.

If anything , given the average level of discourse here, I’d say this is anti-social media.


Exactly, this is not social media. No felony.


It's a site where we browse published stories and then go at length having meaningful discussions. The pinnacle of social media.


According to whom? I subscribe to descriptive linguistics , and I will bet you a good sum of money that if you surveyed a representative subset of English speakers about “social media”, the single most common word in all the answers would be “friends”. Now, please don’t anyone take this personally: I appreciate the content, and I wish everyone here the very best , but I cannot with all the goodwill in the world call anyone here a friend.

It may be different for Reddit or Tumblr or whatever. But I don’t know anyone here by handle (aside from Nagle , Graham , Altman and McKenzie , but only because I know of them from the real world, and they don’t know me). There is no culture of identity at all. Signatures are actively discouraged. Etc etc , long story short: nothing social about this place. No friendships, no social structure , just discussions.

This is anathema to social media. Putting this on the list should be a felony ;)


Who could possibly remember every forum they have created an account on?


But he remembered this one and is saying he won’t list it.


I'll probably have forgotten by the time my visa application comes around.


Better that than be denied.


Contact immigration lawyer, maybe?


I wonder if there is a 'job creation system' at play here for those lawyers to have a job or exist in the first place. If a country has a good and functional administrative system in place, that would not be needed.


This is after all partly why the US tax system is so complex... It's well-known that tax preparers lobby against tax simplification.


ha! I didn’t catch that until you pointed that out. I wonder why is it dated July 27, 2018... maybe the report is still a work in progress?


I had this problem last year, and the only thing I managed to do was forcing my peers to all install TeamViewer :)))

Very specific service that potentially can generates you enough side-income :) I love this kind of project!

If anything, maybe create a freemium model... make it free for anything less than 3 mins?


Thanks for feedback. It is good to see we werent only ones who have faced this challenge :)

Freemium could be good, but at current setup it is not easy to implement.

Still, that would be quite nice feature.


Question here, how do you put a name on the hex? Do you have a database that map each hex code to "limelight", "grey asparagus", etc ?


I am using a database with Resene color names (and some names from the Wikipedia colors list) mapped to specific hex codes. I find the closest hex code from the database to the one in your camera and show the name associated with it.

Hope that answers your question!


You could also use the layperson-supplied colour names from the XKCD survey:

https://blog.xkcd.com/2010/05/03/color-survey-results/

http://xkcd.com/color/rgb/

It took colourblindness into account.


At the very bottom of the page you can see the list used: http://people.csail.mit.edu/jaffer/Color/resenecolours.txt


How else would it be done?


Using a RNN that makes up plausible names based on a training set of actual colors.

I kinda hope that was a genuinely curious question and not an arrogant rhetoric one.


Why RNN? I would probably settle for a nearest neighbour in Lab color space.


Because I'm no data scientist but I somewhat understand that a RNN can extrapolate a training set. Nearest neighbors are cool too!


your conception of RNN as a magic bullet is sorely mistaken.

for something like a meaningful name for a colour requires far more learning, data and context than is reasonable...


Welp


You've got like 3 features, and a list of names with values for each one of those features. You could literally do

    _, result = min([(sqrt((i.R - c.R)^2 + (i.G - c.R)^2 + (i.B - c.B)^2), c) for c in colors]) 
and you're done.

Why would you want a computer to come up with color names anyway? They're identifiers, so you want them to be consistent. What if it comes up with names like Piss or Ennui? Why go through that trouble?


You can skip the square root, too.


INDOSYSTEM | Tech Lead / VP Software Engineer | Jakarta, Indonesia | FULL-TIME / CONTRACT https://indosystem.com

We are looking for senior-level backend software engineers who are excellent with PHP, NodeJS and building system for a mobile-oriented architecture. Tech Lead and VP level position is available for those who want to step up, lead and build the engineering team.

Indosystem is engineering-centric company, currently 20+ team strong, based in Jakarta & Cirebon.


Wow, great comments about Bloomberg. It makes me wonder, is there similar system for Cryptocoins traders?


This is a really interesting idea. Certainly, there's a huge need in the cryptocurrency world with getting information accurately and in one place. And it's only going to grow.


Kind of. It matters for me because it will ruin my video recommendation list.


You can delete those from your history, somewhere in profile settings.


may I know the name of your app? would love to see an example of 5-star RN app :)


sleeperbot


Big fan of sleeperbot, especially the weekly sleeper report. Had no clue it was developed using RN


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: