zephyrnh

If they can make that conversion this seems like it would be a steal. $225/sq ft versus over $1000/sq ft for residential in SF. SF doesn't seem like somewhere that would vote against more housing being available, so if it doesn't even involve rezoning, I'm curious if that's a viable path for this and other office buildings.

If this trend continues, with less office space used, while population numbers stay stable or increase, is there a possibility of commercial buildings like this being rezoned to residential?

> Tech has been bending over backwards for quite a while, but to no avail.

You're applying a false equivalence here. Some things are positive changes and make sense, and some are not. Grouping everything together isn't helpful.

I am male, and my first thought on reading the title was that it was juvenile to a point that I was surprised to see it as a title on HN. Hearing someone else echo that feedback is helpful and will hopefully help the author in their growth.

This is not a "bend over backwards" request, it's a dumb sexual joke on HN, and your response comes across as very condescending.

It's been 12 days since they tested everyone. My understanding is that the results for serological tests are available right away, so I was expecting them to publish results within a few days. Any idea why they haven't yet?

It sound like they're using land that Google already owns for this: "First, over the next 10 years, we’ll repurpose at least $750 million of Google’s land, most of which is currently zoned for office or commercial space, as residential housing. This will enable us to support the development of at least 15,000 new homes at all income levels in the Bay Area, including housing options for middle and low-income families. (By way of comparison, 3,000 total homes were built in the South Bay in 2018)."

"The security lapse was first reported last month, but at the time, Facebook said it only happened to “tens of thousands of Instagram users,” whereas the number is now being revised up to “millions.” The issue also affected “hundreds of millions of Facebook Lite users” and “tens of millions of other Facebook users.”"

Sure, but there's a big difference between a college kid at 20 and a man at 35. I don't know where you draw the line on things you can assume someone has outgrown, or how they've changed, but I draw it at some point where what he said at 20 when facebook was still a Harvard project doesn't affect how I think of him or the company today

This incident is the first one I've felt this strongly about. There have been many others, but even Cambridge Analytica, to use that example, was just taking information that was semi-public (your likes and interests that all your friends could see), and abusing that information. But that was information that I never mentally compartmentalized as private. Sure the scale and method of abuse was unprecedented, but I also don't blame facebook as much as many people did for not really knowing the extent of how that data could be abused. I also believe that fake news spreading on facebook was a novel-at-the-time phenomenon who's impact was hard to detect until after the damage was done.

My private messages are a whole different category of private. Facebook had a phenomenal engineering team and I put the same trust in them that I put in google for my email. A hack is still possible, but it's the highest level of trust that I can have in a service that I can't control (sure things like Signal exist, but 99% of my friends don't use it, so there's a tradeoff). So this particular incident, and the dismissiveness of the response, is my dealbreaker

I assume the point here to send someone a message on FB with a Spotify link, so they click on it in their messages and it opens up the Spotify app. If you just want to send a message from one Spotify user directly to another in Spotify, you don't need FB messages at all, right? Spotify has a list of all your FB friend IDs already and knows which Spotify accounts each is connected to

I assume someone at Facebook, hopefully the person that wrote this, or someone who has more influence over this issue, is reading.

I am an engineer. I understand technology better than most of the general population. When I sign in to my Facebook account to use Spotify, I am absolutely not expecting that Spotify will now have access to read every single one of my private messages. This is a gross violation of trust, and if this is what happened, then the fact that you not only made this mistake, but also then published this blog post defending it, marks a low point for Facebook. Perhaps irrecoverably so for me.

"After signing in to your Facebook account in Spotify’s desktop app, you could then send and receive messages without ever leaving the app. Our API provided partners with access to the person’s messages in order to power this type of feature."

This is a write permission. So you needed to give Spotify permission to create a message. It seems that your system combines the read and write permissions, since you just grouped them together by saying "access to the person's messages". It also seems from your defense that you see absolutely no issue with this. In order to share a song through Spotify, you are giving them access to every single private message the user has ever written.

I find it hard to believe that Facebook refuses to acknowledge any fault in this: The initial product decision, the upholding of this decision through previous privacy investigations, and this PR response. Am I misinterpreting the facts or scale of this?

I don't know what I'm voting on this yet, but it's certainly not a simple matter, and there are many smart and compassionate people who don't support the measure. Here are the mayor's thoughts on homelessness (https://medium.com/@LondonBreed/a-bold-approach-to-homelessn...) and why they oppose the measure (https://medium.com/@LondonBreed/statement-from-mayor-london-...) As I mentioned I don't know which side I'm on yet, but to just lash out at someone who opposes the measure as "disgraceful" means you're either uninformed or unwilling to consider other reasonable perspectives as legitimate. Either way, you can do better

Airbnb switched to a "double blind" review system several years go, so you only see the other person's review after you've written your own. If you don't leave a review within two weeks, the other review becomes public and you no longer have the option to leave a review. I did a search to see if there was any press about this, couldn't find much but there's a bit about it here: https://medium.com/airbnb-engineering/building-for-trust-503... "The result was a 7% increase in review rates and a 2% increase in negative reviews."

Exactly. bayesianhorse's approach is perfectly valid, but your response describes our reasoning for our approach.

We'd love to do this. As you can imagine, however, it's a very time consuming task, and there are a lot of competing priorities (including other projects we've open sourced) and we therefore can't make any guarantees about if/when we'll be able to do it.

Agreed, this describes our reasoning pretty well. Additionally, as we mentioned towards the end of the post, the incremental benefit of a model not supported by PMML is unlikely to be more significant than the incremental improvements we see from investing in improving our features and ground truth. As such, our lowest common denominator isn't actually the model, but rather the features and data pipelines. The suggestion sytelus makes is a perfectly good way of doing it though, and we will likely change our approach when we find that our models do become the lowest common denominator, or have a higher relative ROI for the time we invest in improving them.

No, features are definitely a business secret. My goal was to share as much as possible that might be helpful to others combating fraud, without giving anything away that would hinder the effectiveness of our systems.

Thanks! It varies depending on the model, but in most cases, yes, the confirmation of fraudulent activity is manual since having the ground truth be correct is critical

"They generate revenue very early on in their lives. Often as soon as they get their first user."

Is this true? If "super successful" can be understood as "the biggest tech IPOs of the last 15 years", then I think that Google, Facebook, LinkedIn & Twitter would be at the very top of that list. I guess it depends on how we define "very early".

Hey everyone, I'm one of the engineers who worked on Verified ID. I'm going to keep reading all the comments people leave, but rather than respond to all of them on here, I've written up some of my thoughts on our blog: http://nerds.airbnb.com/verified-id/

I'd just like to stress that this doesn't address all the issues brought up, since some of them require more time, but I'm happy to answer more questions if you leave a comment on my post.

Thank you all for your response!

Thanks for sharing your thoughts - RE: Talks not being local, I agree with the comment below about livestreaming.

RE: email, we do have an email list. However, we really don't want to spam people, so instead of opting you into it automatically, we make it opt-in here: https://www.airbnb.com/techtalks.

It sounds like we may have err-ed too much on the side of being non-spammy, and this list is not very discoverable. I still don't want to automatically opt attendees in to the list, but given what you've said I think we'll pass around a signup sheet at the next tech talk.

Thanks for the feedback - I work at Airbnb, and I'd be interested in hearing if others share this feeling (or don't).

Yeah as far as I know, the current system is just a bunch of eventbrite links being tweeted/shared for different events. Would be great to have it centralized somewhere, especially if done in a good way that allows filtering out talks by focus area, how appealing people find them (upvoting?), etc.

I can't speak for Yehuda, but based on the talk abstract (in the link), it doesn't seem like ember will be discussed.

San Francisco, CA H1B Welcome

Airbnb - http://www.airbnb.com/

We only have 27 engineers, and we need more!

I'm a backend engineer here, and we're hiring backend, frontend, mobile and ops engineers. Check out our (pretty sweet) jobs page: http://www.airbnb.com/jobs

I've been working at/founding startups for the last few years, and didn't think I would end up somewhere as "big" as Airbnb, but I love it here. Despite the large size of our customer service operations worldwide, the entire product team (PMs + engineers + designers) is still ~40 people, there's still a lot to do, and we need help doing it :)

*of=or

Not sure if you mean Airbnb tech talks of Yehuda's past talks. Some of Yehuda's previous talks can be found here: http://yehudakatz.com/tags/conferences/

Previously recorded airbnb tech talks will be posted here this week: http://www.airbnb.com/tech_talks

Yeah, sorry about that, realized my mistake after I submitted and seems like I can't edit the title (information in the link is correct). Upvoting your comment so people are aware.

FYI This talk will be video taped and streamed - we'll have links for both up later.

Sorry for the late response - was looking into this, and unfortunately this talk won't be streamed or recorded. We hope to record future tech talks, and will make it clear whether or not they're recorded in the future (we have them every two weeks).