HNHacker News
TopNewBestAskShowJobs

statusgraph

159 karma · joined October 8, 2012

submissionscomments
statusgraph··on Why Sweden has so few road deaths
This is normal in the US for roads in mountains (going uphill) and it generally works great.
statusgraph··on What are the lesser known but useful data structures?
An append only log? The value of writing data is that you can read it, otherwise your structure is semantically equivalent to not writing at all.
statusgraph··on Requirements for DRM in HTML are confidential
It's pretty unreasonable to claim fraud (charge back) if you buy a game and then someone gifts you...
statusgraph··on Find Friends Abuse
One typically uses salting to randomly generate a hash function. (so even if you know it was, say, SHA512, you have to guess the salt to reverse engineer)
statusgraph··on Find Friends Abuse
I'm not certain what jluxenberg is proposing with the hashes. But one possibility is that the exposed API will accept hashes that only the app can generate (through a salt). Although obviously you can crack the app, I suspect this is significantly harder.
statusgraph··on Snapchat Checker
It probably did, but the site is attempting to provide some level of privacy.

Interesting decision. Sure the dump is publicly available, but this is much more accessible.

statusgraph··on Watch pg write an essay in real time
Are you sure? It feels more like edits are being played back at a constant rate, without consideration of human delays.

But still super interesting

statusgraph··on Your body wasn’t built to last: a lesson from human mortality rates
By that definition all laws of physics are laws of observation.
statusgraph··on Wolfram Language
They've existed in a significantly more powerful form as DownValues since Mathematica's release.
statusgraph··on Google Wallet Users Can Sign Up Now for Physical Cards
Credit cards charge merchants for their loyalty programs. They do make money from carried balances, but that's not how reward programs are funded.
statusgraph··on How I Learned to Work a Room
Your point of exiting negative chats is a very good one. The instinct (for me, anyway) is to double down and try to fight it out, which is a waste of time for everyone involved.
statusgraph··on How much revenue top companies earn per second
Why is customer supporter log(n)? Seems it would be nlog(n) (because of management overhead)
statusgraph··on The Hidden Job of the M7 processor in the iPhone 5s
Not applying filters is precisely the right thing to do; it gives developers the most flexibility in interpreting the data. Filters by definition lose data, and can only be chosen with a specific use case in mind.
statusgraph··on How much can an extra hour's sleep change you?
Probably a pretty girl in his first class :)
statusgraph··on Heroku WebSockets Now in Public Beta
They're very different from typical HTTP requests: very long lived and generally cannot be buffered.
statusgraph··on After 180 Websites, I’m Ready to Start the Rest of My Life as a Coder
Haha, not sure that qualifies as a "nit" given that it invalidates their point entirely ... :)
statusgraph··on Ask HN: Who is hiring? (October 2013)
Square — NYC https://squareup.com/ ---

Square! In October 2012, we acquired the design firm 80/20 and have had New York based colleagues for some time now. Our logical next step was to bootstrap a NYC Engineering presence to tap in the growing community here. We just started building the Engineering team in July.

As a company, we have four main focuses.

Go international. We're adapting what made us successful in the US abroad. This is harder than typical international plays since it requires translating our four major benefits to small merchants: 1. Shipping readers, 2. Card payments (physical world is much more complex than online), 3. Sending money quickly to our merchants, a.k.a. settlement: very localized and next-day is hard on a number of fronts), 4. Assessing risk of merchants as they sign up, aka underwriting: again, very localized most countries do have FICO scores, SSN, D&B numbers and such.

Move up market. One of the biggest success of Square is to have enabled merchants who were "non-consumers" to start accepting credit cards. In classic disruptor fashion, we have essentially alonged the long tail. We haven't been eating someone else's lunch, we've created a whole new category! Now it's time to move up to bigger merchants.

The other side of the counter. With Wallet, Square Market (http://squareup.com/market) and Square Cash (http://squareup.com/cash), we're going on the other side of the counter and interacting directly with consumers. We think that we can provide the best experience for consumer-to-merchant and consumer-to-consumer commerce. We're at the infancy, but extremely well positioned to succeed.

Data as product. We're leveraging our data. We know a lot about our merchants: daily sales, seasonality, growth of their business, correlation with other events (e.g. Mapping Square Payments During the Super Bowl). Lots of innovative products are in the making which will highlight the breadth of what we do.

In NYC, we're starting with a focus on infrastructure, distributed systems, and money moving systems. We're a small team, located in SoHo, and defining the culture of engineering for Square on the East Coast.

If you're interested; I'd love to chat. Shoot me an e-mail with your resume! wkm@squareup.com

statusgraph··on Apple Updates iMac
Do you mean MacBook Air? The Pros have been retina for a while :)
statusgraph··on Square open sources Kochiku
It sounds like Square doesn't use 10 mac minis but rather EC2 spot instances....
statusgraph··on The Banned Hashtags of Instagram
Exactly; same same with "fuck"
statusgraph··on Why is FreeBSD deprecating GCC in favor of Clang/LLVM?
This is exactly GCC's architecture as well, where front-ends are transpiled into a generic language, which is then compiled into native code for a given platform.
statusgraph··on Billing Incident Post-Mortem: Breakdown, Analysis and Root Cause
Interesting to slave redis across datacenters. I wonder how well that works in practice.
statusgraph··on The Real Value of Stock Options
What this analysis shows is the flaw of trying to reason with percentage: the modeling is too difficult and littered with assumptions.

An alternative model: say you're offered $100/yr with 10k options a year with a nominal value of $5 and a strike price of $1. Superficially the offer is worth $140/yr. The startup's valuation is still important, because you can ask yourself questions about how likely they are to double it. It's much easier to project if the company can double to 10x it's value than what their exit is going to be.

Say that startup is worth $1mm. If they can double it to $2mm you're going to be netting $100 + 10k(10-1) = $190k. But can they double it? Up to your impression of their business. That's a much simpler question than wondering about their exit many years down the line. (their competitors generally are the strongest signals)

Observations:

this is computed with options per year; most startups would put this as giving you 40k options over four years * the options companies give you generally have expirations and very poor liquidity so using their face value is generous. [1] * but on the other hand, options can potentially increase in value, so it's not wholly unreasonable to equate them to their face value * this approach cuts out nearly all early stage startups which give below-market salaries and weak equity grants. Getting 50bp of a $10mm startup over four years is $12.5k/yr face value. If market salary is $100 and they're offering $75 you're netting around $87.5. That's a pretty stiff cut. * dilution isn't relevant (assuming you have options on reasonable stock) * if the startup is doing well (say, after a year doubles their valuation) it's difficult to negotiate another grant because you're now paid quite a bit better. Conversely, if the startup is doing poorly time to ask for more. * you may be okay with the pay cut because of externalities (more interesting work, good for the resume, better commute, etc.)

(ps. I normalized all salaries to $100k because it's easy to do percentages off of. Adjust by industry and specialization.)

[1]: That said, if you have reasonable timing you can use this computation to negotiate comp when changing companies so you're not completely handcuffed. (basically converting your equity from one company to another)

statusgraph··on Stunned by Go
It doesn't work :) http://play.golang.org/p/PVS5sarmHb
statusgraph··on Somewhat Verbatim Conversation with a Startup Recruiter
I get your point, but the difficulties startup founders have to endure are irrelevant to the value proposition they make potential employees.
statusgraph··on Erlang at Basho, Five Years Later
The gocircuit project is trying to make distributed goroutines practical but it seems a little rough: http://www.gocircuit.org/
statusgraph··on Let's deploy via Git
Note that delay-updates is not actually atomic [but it's closer] =)
statusgraph··on Oculus Rift co-founder killed by gang trying to escape police
I agree with you, but I think most bus factor considerations are about poaching/the developer leaving =)
statusgraph··on User Motivation Determines the Best Color Scheme for Your Website
White on black is very hard to see in a brightly lit room. Many programming talks like that color scheme and it just doesn't work with crappy projectors.
statusgraph··on How we send a morning email to users around the world.
my personal favorite is the 45minute offset: http://en.wikipedia.org/wiki/Chatham_Islands#Chathams_Time
← PreviousPage 2 of 3Next →