HNHacker News
TopNewBestAskShowJobs

sshumaker

1,535 karma · joined October 16, 2009

email ===== work: scott at persona-ai dot ai personal: sshumaker at gmail

Product-focused tech executive with 20+ years experience building consumer and enterprise products. Currently co-founder of a stealth startup building something outrageously ambitious.

Previously VP at Microsoft and at Credit Karma. Ex-Googler and former gamedev.

submissionscomments
sshumaker··on Laying Out iOS UIs in Code
No offense to the parent, but I think AutoLayout is kind of flawed. It's a good idea in theory, but has some very rough edges in practice. When things go wrong, it's difficult to figure out why, because all the constraint solving happens elsewhere. You might have an over or under constrained layout and end up with problems. In our experience, it also extracts a decent performance hit for complicated layouts (tossing hundreds of rules into a constraint solver is not fast).

Some things were harder than they needed to be, as well (e.g. centering a group of vertically layed-out elements). We often needed to create spacer UI elements to work around it.

FWIW, I rebuilt an iOS5-compatible version of AutoLayout (using Cassowary for solving and a lot of objective-c runtime magic to maintain a shadow hierarchy that consists of layout objects), which we even shipped for a while. Eventually we decided to move away and write our own system, with an emphasis on debuggability and simplicity. We ended up with a system that kept 90% of the benefits of AutoLayout but without needing to solve constraints.

The key element was the idea of 'smart rectangles' which were used for layout. You can set any two components explicitly of an axis (e.g. left and right) and the others (e.g width and centerX) would be determined automatically. Basically each rect system remembers the last two components on a given axis.

We also have the concept of a selection, which takes a group of 'rectangles' and can perform various operations as a group, like laying them out vertically, centering them all on an axis, and moving them as a group. This makes it easy to do things like center a group of vertically layed out elements (select, layout vertical, then set the selection's centerY).

The rectangles can then be applied to UI elements (which automatically handles rounding to pixel integrals, and ensures you don't end up updating UI unnecessarily as you build your layout, which is the problem with most of the UIView categories). But you don't necessarily need a UIView to apply - we've used this system for laying out OpenGL elements as well.

In practice, the code feels kinda like Autolayout:

  buttonRect.centerX = bounds.centerX;
  buttonRect.width = 100;
  buttonRect.top = widgetRect.bottom;
  buttonRect.bottom = bounds.bottom;
Unlike AutoLayout, you need to lay out stuff in order (e.g. you in the example above, widgetRect.bottom has to be set already before you assign across to buttonRect.top), but that was a small price to pay since you can step through the debugger and instantly see any changes you're making if your layout doesn't work out. And we don't handle constraining between non-sibling views as nicely as AutoLayout. But for simplicity, performance, and maintainability, it's been a huge win for us.

Hopefully we can eventually open source it - all of the devs who've used it miss it when working on personal projects.

sshumaker··on Ask HN: Best big companies to work for?
This policy varies on a team-by-team basis, but generally Google is pretty good about giving people some flexibility in this regard. It's pretty common to see people work from home because they e.g. have a dentist appointment in the middle of the afternoon or something. We generally treat people as responsible adults unless they give us a reason to otherwise.

That said, we really value the team-building and communication that happens when everyone is in one place, and that typically means WFW (work from work) the majority of the time, so we try not to make a habit of WFH.

This is kinda the same issue with remote work. We have great videoconferencing tech, but we generally find that you miss out a lot when you're apart from the rest of the team. This makes even being on a geographically distributed team kinda difficult. Some teams do a better job than others on this (e.g. Chrome is very far-flung), but it requires a lot of coordination and discipline to make it all work.

sshumaker··on Ask HN: Best big companies to work for?
Read "The Algorithm Design Manual", and do exercises from every chapter.

Whether or not you decide to interview at Google, building up strong CS fundamentals will make you a better programmer. And we're not the only ones who ask those types of questions. :)

sshumaker··on Ask HN: Best big companies to work for?
I have to disagree. Just in the last few years, tons of new, groundbreaking projects have sprung up. e.g. most of the publicly announced Google X projects didn't even exist a few years ago. And there's so many areas where there's interesting stuff to work on.

Now obviously, unless you're really special (e.g a known expert in your field) you're not going to be able to waltz in and be given a huge area of responsibility and a team, but that's just common sense. You do have to prove yourself, and there is some discounting down of your work pre-Google (your impact since Google is valued far more highly).

At the end of the day, Google is a heavily engineering-driven company, so if you have good ideas, nothing is stopping you from going ahead and starting your own cool project. If you're driven and talented and demonstrate success, the 'cool project' could transform into something much bigger, and you could end up leading an important company initiative.

sshumaker··on Ask HN: Best big companies to work for?
Even though you're expecting it, I did want to mention Google. We're actively hiring and there are lots of really great opportunities, especially for startup / mobile folks who also have really strong fundamental CS backgrounds (e.g. algorithms, data structures).

Having come from the startup world myself, I'd encourage anyone who wants a change of scenery, even if it's only for a few years, to apply here.

  * Great company culture, competitive pay, and amazing benefits
  * Lots of interesting problems to work on
  * Learn about developing at scale
  * Good work-life balance. 
And, especially if you want to go back and do a startup again in a few years:

  - Expand your network of really talented people.
  - Understand the kinds of products and teams the top-tier companies are looking for
Although I also wouldn't be surprised if you decided to stay for a lot longer.

One of the downsides to the hiring process is that team placement can still be a bit random. If you're a senior dev who still loves actively coding and is in the startup world, and is thinking about applying but is worried you'll end up working on something that doesn't interest you, drop me a line at the email in my profile. Especially if you're interested in mobile.

Disclaimer: I'm a hiring manager at Google.

sshumaker··on Ask HN: Best big companies to work for?
The hiring process at Google is affected by a lot of variables. But for senior engineers, the salaries + benefits (especially stock and bonuses) are definitely very competitive.

It's also much easier to negotiate a better deal if you have a competing offer from another firm. And doing extremely well on the interview process also helps, along with a strong track record or in-demand skill set (e.g. mobile development). This is generally applicable to the most tech companies, not just Google.

Disclaimer: I'm a hiring manager at Google.

sshumaker··on Google's vs. Facebook's Trunk Based Development
Au Contraire. Google (and I'm assuming Facebook) have very detailed dependency rules that are explicitly described in files in nearly every directory throughout the codebase. There's no way you can do massive cloud compiles without this kind of information.
sshumaker··on If Steve Ballmer Ran Apple
Definitely doesn't seem that way from the inside. We're constantly productizing new research that comes from R&D. Self-driving cars, Glass, DistBelief, etc.

In fact, one thing that's unusual about Google is how much R&D actually makes it into products (unlike, say, Microsoft, whose research group turns out great stuff but rarely does anything with it). In that sense, it's less pure blue-sky, but more applied. I think it's more akin to the Manhattan Project, Apollo program, etc - scientists and engineers working closely together to redefine what's possible. (We even internally refer to the ambitious projects as moon shots). SpaceX and Tesla are doing work in a similar vein - an virtuous cycle of innovations and research.

sshumaker··on Apple Acquires Nigerian Tech Entrepreneur's Startup HopStop
And who haven't already switched to Google Maps...
sshumaker··on You know, Google, the web already had this feature
http://news.cnet.com/8301-1023_3-57413309-93/larry-page-goog...

"[Larry Page] ... called [G+] the "social spine," reminding listeners that the service isn't just a destination, but a unifying social framework for all Google's products."

First off, Google allows you to have multiple identities with SSO. And I find it hard to argue that it's not a good thing for 99% of the user base. Managing multiple accounts sucks, and being able to share between them (and supporting a very rich set of privacy tools) is a win for most users.

A genuine question: Why don't you want your YouTube account linked to your Gmail?

sshumaker··on You know, Google, the web already had this feature
I think you're fixating on Google+ as a social network, when that's not really the point. Google+ is meant to serve as an identity layer for Google, so you don't have a bunch of different logins / passwords to silo-ed properties, which makes a frustrating user experience. They give you pretty damn good tools for deciding what parts of the identity are shared for this exact reason. You can enable Google+ and basically ignore the 'destination social network' entirely if that's what you want.

As for having Play store comments attached to your G+ identity, what is your complaint there? I'm sure the quality of reviews is way higher when you can't hide behind an anonymous account.

sshumaker··on Joe Armstrong: Solving the wrong problem
Ultra-lightweight actors, a VM scheduler tuned for scheduling massive numbers of concurrent ops, etc. If you tried doing this in with a java library running java code, you couldn't get anywhere near their level of concurrency.
sshumaker··on I’m done with the web
If you actually bothered to read his post, he was actually using ember.js as his example, which is decidedly not written in Objective-J. It's plain javascript with handlebars templates.
sshumaker··on Amazon stops selling Sim City V over game issues
That's slightly revisionist history. EA bought Maxis when they weren't in great financial shape (it didn't help that basically every Friday was spent at the bar). It's true that the Sims was a huge success that EA didn't anticipate (they repeatedly tried to kill the project), but the jury is still out if Maxis would have survived long enough to release it without EA's funding and support.
sshumaker··on Letter from the grave (2009)
Who is the power behind the throne the author is talking about? Is it Gotabhaya Rajapaksa, the president's brother and minister of defense?
sshumaker··on 75-year-old soybean farmer sees Monsanto lawsuit reach U.S. Supreme Court
They do own technology to achieve this: http://en.m.wikipedia.org/wiki/Genetic_use_restriction_techn...

It basically prevents G2 seeds altogether, but they've pledged not to use it because of public outcry, instead relying on these contracts.

If you seriously think through the consequences of usage of this kind of tech, it's pretty dire. Due to natural cross-pollination, we could be setting ourselves up for a world where everyone must buy all of their seeds from Monsanto and their ilk, who would have a monopoly on the world's food supply. That's bad news indeed.

sshumaker··on Google's Driverless Car Is Worth Trillions
Evidence?
sshumaker··on U.S. Attorney Carmen Ortiz Issues Statement on Aaron Swartz Case
Pollard has been transformed by opportunistic Israeli politicians into some kind of Israeli patriot. He's not. He did it for the money, selling secrets not only to Israel but to Pakistan (!) and South Africa as well.
sshumaker··on Google's text-to-speech engine randomly inserts phrase "he now praises the ipad"
Looks like the 'filled with so much drama' quote is from here:

http://www.macnn.com/articles/11/12/01/surprised.that.reader...

Strange, it appears to be text from a random news article. Some kind of bad training data? Or memory corruption?

sshumaker··on A new kind of Humble Bundle
What Double Fine is doing here is really clever. Publishers are notoriously risk-averse, and by demonstrating interest for a game concept, they can more easily fund innovative games.
sshumaker··on LuaJit 2.0 is out
Also the addition of the ffi library. It totally changes the way you integrate with C code (for the better) - its the best FFI I've seen in any language.
sshumaker··on Why Google+ Still Has Huge Potential
Just curious - what do you find so distracting?
sshumaker··on Google offers tool to bridge Android and iOS app dev
Speaking as someone who has developed on both platforms, it's definitely my experience. And that includes starting with an Android app and porting it to iOS - half of the stuff I had to add to Android to make it usable I discovered they had already built into iOS. Not to mention the poor choice of defaults, coping with device fragmentation, and the difficulty making Android apps look attractive via-a-vis iOS.

I haven't built ICS apps, though, so perhaps it's getting better.

sshumaker··on Google offers tool to bridge Android and iOS app dev
Well, that's the point of the java and objective-c bridge: it lets you call out to native APIs, so you have all native components.
sshumaker··on Google offers tool to bridge Android and iOS app dev
Not to mention the iOS SDK is much nicer.
sshumaker··on Google offers tool to bridge Android and iOS app dev
Toss in lua(jit) and you have a nice dev environment. I wrote a lua->java bridge and modified LTC (lua tiny cocoa, a lua objective c bridge) and with those you can write nearly your entire cross-platform app (ui, too) in C / Lua.

And with luajit's FFI support, it's almost a better C than C.

sshumaker··on J2objc - A Java to iOS Objective-C translation tool and runtime
You can already write code in C/C++ that works across both platforms. It's trivial to interface with objective-c, and pretty easy to call from Java using JNI.
sshumaker··on Greg Knauss's 10-year-old son's $23,800 bug bite
You mean like auto insurance (liability)?

You're subsidizing risk across a large group of people. The chances of something happen to an individual are small, but could be financially devastating. So everyone pools their risk together for affordable premiums.

sshumaker··on The U.S. Senate has blocked the Cybersecurity Act of 2012.
Well, except for the fact that they're heavily bankrolled by über-wealthy conservatives (Koch brothers, etc).
sshumaker··on The Silent Majority of Experts
You're wrong.

There's only a mild correlation between fame and skill set. I can name a few dozen 'famous' internet developers, and have even interviewed a few of them, and most weren't anything special, other than being good bloggers.

Meanwhile, take someone like Jeff Dean (who I just learned about a few days ago). Designed and implemented BigTable, MapReduce, Protocol Buffers, etc. I'm not sure most people outside of Google have even heard of him.

Sometimes your work finds the limelight but you aren't credited for it. We all know Elon Musk but who knows the guy who actually wrote the code for the rockets? I don't remember the name of the Amazon Dynamo guys, and have no idea who designed EC2 and S3. These are awesome systems used by millions of people, but the creators are (relatively) unknown.

← PreviousPage 6 of 11Next →