erdevs

Connecting something like this to github and other large public repos would be fun.

Similarly for pseudocode rather than only ins/outs.

Maybe code analysis for similar implementations to (or optimizations of) what you're in the middle of developing.

Have long wondered about this, but never got around to dinking with it at all. Cool to see things along these lines.

If accurate, the description of British moral culture at this time sounds abhorrent. I wonder when and how it evolved to be more compassionate toward the less fortunate.

Of the two, between math and code, code is the more obfuscated form of the pair. Can not most code be expressed as pseudocode? Most pseudocode as an algorithm? And algorithms as math? Math is the purer, more fundamental, less obscured form. What we do with code is adapt and elaborate the unobfuscated essence of the underlying algorithms in order to make them rotely executable by machines, and we further twist and obfuscate the essence in order to optimize the coded algorithn for its bespoke execution environment. The math underlying a given piece of code strips all of this away.

Aaronson is awesome. Intellectual honesty like this is to be admired in any field.

I wish that more of silicon valley would penalize people for reprehensible behavior. If Uber execs were ostracized rather than lauded... if investors pounded the table about legal risks and the importance of ethical business rather than cheering along any behavior that makes the numbers grow (temporarily)... we might not have Zenefits melting down, nor sociopathic exec teams at major companies that are held up as scions of the tech world. If this is one of the best and brightest companies the tech industry has to offer, we have a deeply worrisome problem. If pursuing paper growth absolutely trumps ethical concerns, we are feeding into a horrible system. If nobody calls this shit out and there isn't actual financial and social/cultural consequence, things will only get worse.

The article asks "can possessions stand in the way of fulfillment?" Experience tells me the answer there is certainly yes, they can and often do. I think achieving a cultural shift away from the accumulation of possessions is achievable-- as cities get more dense, as travel becomes more regular, safe and affordable-- and in some sections of culture this is happening already.

The article also asks though if we can truly ever own things vs only custodian them. Here I don't see any shift happening. The principal of property ownership, where property extends to land and to ownership of business corporations, is what largely drives the Pareto distribution of income we experience. This, in turn, defines the disparity between the rich and the poor. So, I'm not sure a truly fundamental shift can or will happen here.

I'm curious to understand the forthcoming results and learnings from experiments in Universal Basic Income which are beginning to get under way. And I think that if robots and AI actually automate away most current human labor, we might be able to rebalance our economy in a manner beneficial to more people. Alternatively, if life extension becomes more and more practical, the fundaments of our economy may be ripe for change. So, over the course of decades or centuries, I can see the equilibria of our economic system and the distribution of incomes and therefore wealth accumulation changing dramatically. But I don't think it happens through any mere cultural shift in thinking nor via incremental policy changes. That may not be a bad thing, but everyone should know that the fundamental rules of the system are unlikely to change dramatically any time soon, without some major impetus to necessitate it.

Well... several possibilities.

Could be that they are lying about revenues being small and just trying to squeeze you. That would be a bad thing.

Or, could be they're telling the truth and revenues are struggling. That is a bad sign in terms of financial health.

Seeking to cut costs when times are hard can be a sign of responsible discipline or a sign of desperation.

It's usually a bad sign that a company offers "great pay" and then has to dial it back. Generally seems like a lack of discipline and financial forecasting savvy and controls.

It's hard to say what is going on here specifically. For most startups, you're dealing with inexperienced leadership and most often these measures are a very bad sign. But this may be an exception.

How much experience do the founders have running companies? What % of pay did they ask you to cut? Can you ask if any layoffs are happening? Can you ask how much cash the company has, what monthly cash burn is and/or how much runway is left?

You should ask for something in exchange for the reduction, if you take it. If you believe in the company, you could ask for equity. You could ask for an agreement that they give you at least, say, 8 weeks severance if they let you go (since accepting the pay cut will eat into your ability to save and leave you in a bad position if they need to lay you off).

I'd brush up the ol' resume and start putting feelers out there, just to be safe! Try to ask some of the questions above and also think of what you'd like to receive in exchange for taking less cash pay.

Good luck!

I think a little personality from the products/services you use is nice. And some personalities are cutesy. As long as it doesn't get in the way too much, I don't see what there is to fuss about.

Personally would rather take what we have now in terms of product voice than revert to the souless, bland, offend-no-one product and corporate voices common as of, say, the 80s.

I doubt the business policy is that you can't take partners out for meals and/or invest in relationship building via meals and entertainment expenditures. Nor is the policy likely that you cannot participate in such activities when offered by the partner.

The key is that you should not exchange company resources for what may be perceived as or actually be personal benefit. Outside a contractual term, set up a quarterly dinner to strategize, synchronize, assess, etc where the vendor pays? Sure! But, explicitly concede a material business term in a contract in direct exchange for a term that personally enriches you? No!

It is enjoyable to negotiate with people who view "concession" in this sense. You can get them to agree to terms of greater business value in exchange for terms of smaller business value, and they don't even think they "conceded" a thing!

It's a concession. Hootsuite didn't want to cover the credit card fees. But they ended up doing so. There wasn't even a proportional split (eg "let's each cover our portion of the credit card fees, according to our revenue share split") or a split of any kind. There was a separate concession by the vendor to cover steak dinners (of lesser value, and only a fixed cost) in exchange for Hootsuite's credit card fee concession... but that does not somehow make Hootsuite's concession a non-concession.

In any case, no point in debating terminology. Hootsuite gave up more value than they gained, but both parties got the deal done and are happy with it, so they made the right call in the end (conceding vs blowing the deal up). It's just not an amazing or widely applicable negotiation technique and, if anything, the vendor likely employed the better technique here.

The thing that unstuck the negotiation was conceding to the disputed term (agreeing to pay the credit card fees). The "steak dinner" trick seems to have little to do with it.

Basically this boils down to "if you're stuck in negotiations but really want to wrap up the deal, consider just conceding on the final, minor points in contention. Maybe try to get something small in exchange for your larger concession(s)."

Not exactly earth-shattering and the hubris in even glancingly comparing this to the brilliance of Nash's work is astounding.

On top of that, this counter-concession doesn't even scale up with increasing revenue. But Hootsuite's fees do increase as revenue increases.

I think it'd be more interesting to read a negotiation tip from the vendor in this situation. Something like "if you've already worked out the major terms in a deal and you are fairly confident the counterparty wants the deal done, but you are stuck on a final point that affects you more than them... consider standing firm, even to the point of obstinance. Test the limits and see if you can get them to concede. If they're holding out, see if you can offer something trivial in exchange for their concession or if you can get them to make an offer to concede in exchange for some trivial ancillary benefit. Sometimes this works as people are irrational and just want to feel they 'gotcha', even if what they're conceding is of greater value -- and has the potential to be of far greater value-- than the exchange. All the better if you can make them feel like they're oh-so-clever in the proposal. We once had a partner we were negotiating with agree to cover credit card fees which would've hurt our net revenue share substantially, simply in exchange for agreeing to buy them a steak dinner once a quarter, which we likely would've done anyway to keep this important partner relationship moving forward and growing over time. Ha! We made sure to cap the price of the steak dinner and merrily signed, having won the business concession we needed."

Besides not even being a great or meaningful technique in practice and certainly not defining any new theoretical work of import (unlike Nash), this is also mildly unethical and would reflect poorly and/or get the practitioner in trouble in many, if not most, situations.

It is sad to see the comments on Medium cheering "brilliant, man!" and "love this."

The headline for this post is far too broad. More accurate would be to say that "very early stage startup technical diligence is usually a waste of time". That is often the case, for many of the reasons the author outlined.

Later on in a company's life, however, technical diligence becomes very important. If anything, VCs should do a better job of it. Countless companies have struggled, stalled or failed because they didn't become aware of or address technical issues early enough and scaled the business side too far out in front of what could be managed technically. This gets your company upside down fast, and in very difficult to unwind ways. Customers become dissatisfied at instability and lack of rapid improvements. Internally, strife between revenue owners and technical owners can easily increase as tech tries to balance delivering new funcitonality and addressing old technical debt.

After you find product-market fit and start really scaling-- ie after the early stage-- technical diligence should be a much higher priority than it currently is, for both VCs/investors and for company operators within their own orgs. Better technical assessments of issues and opportunities would serve most startups very well.

This was my take exactly. Also, the strategy seems ineffectual, focusing mostly on influencing discussion communities when the real rub with Bitcoin has been disagreement among the core devs. If someone's goal was to stop or slow Bitcoin, you'd think they'd focus their strategy on at least trying to shape the behavior of core devs and trying to drive wedges among them.

Are you sure? This company sells Windows-based security software. While we can't speak definitively to their intent, it is possible to objectively observe that they stand to gain from the impression that Windows is less secure (and more in need of security software like theirs).

I had the same thought exactly. :)

We need a new term for "unicorn" ... "Paper unicorn", perhaps.

A better comparison would remove the confounding variable of valuation. Instead, it'd be interesting to look at start-ups in each region with >$200M in annual recognized revenue. Valuations fluctuate based on investor sentiment and market conditions. Revenue is more fundamental.

Yes, but that's kind of my point. I was replying to people speculating that SF must have more than ~526 people with liquid net worth above $30M.

526 doesn't seem a low estimate to me. Unless your counting unrealized paper Gaines... Before tax, and which often do not pan out upon liquidation the way a naive reading would imply they will on paper.

There is an enormous difference between paper money and real, liquid money. There is an enormous difference between the headline price an investor invests at, with all their Preferred rights, and what an employee can actually cash out at.

There are very, very few cash exits of $1B+ in any given year. And far less than the majority of those have rights that allow employees, or even founders, to cash out at, for example, the sky high initial IPO market cap pop price.

Moscovitz did not say that employee #100 made $200M at FB. Read the article. First, this was an overly simplistic analysis. It assumes a strike price of $0.00... Which never happens, and certainly wouldn't have happened by the time FB hit 100 employees. It assumes full vesting. It assumes no lockup on liquidity. It assumes no dilution from exiting... which, in the real world, never happens.

You can actually see how much money FB's top shareholders were worth at their IPO. Very, very few employees made over $10M. TEN. Before tax.

A lot of people front in SF / the Bay Area. Very, very few are worth over $30M in liquid net worth.

Perhaps another takeaway from this article is that it'd be nice if more research papers benchmarked with vastly larger data sets.

Many researchers would love to do just that, of course. However, as many researchers will lament, it's not always easy to get 10+-figure node and 11+-figure edge data sets appropriate to a space being explored.

The best research benchmarks I see do compare to a single-core (often multithread or multiprocess) implementation. And they also show benchmark results on datasets of increasing sizes.

I agree with the authors that those sorts of papers aren't common enough, though. And we should strive to do better. Moreover, I agree that in practice in industry, many people over-optimize for horizontal scalability early, and/or do not realize potential savings and benefit by doing vertical optimizations after gaining initial scale.

I'm surprised the author doesn't ponder the following: Perhaps higher IQ means a person has more ability to solve these sorts of problems, if they take the time to exert the effort. That ability doesn't guarantee success in these sorts of questions without applied effort, however.

That would mean IQ is still very, very important. But that it is not the only importans factor in solving problems well (eg rationally).

To me, that seems quite plausible. Not sure why the author takes a rather anti-IQ slant as opposed for promoting supplemental measures.

This was a great post. Having seen several apps deployed in the app store myself, it seems to me OP is right on that discovery is bad (beyond editors' features, search and recommendation systems are terrible). It makes it really hard to gain traction through word of mouth. Nowadays, especially for a game, you need to plan on having a large UA budget.

Google Play seems better than the App Store right now in terms of both search and recommendations (though both are still bad). Still, it is brutal out there and you need not just hefty UA budget, but a real operation around this stuff.

OP had great thoughts on solving this. App Store / Play APIs would be awesome. Wonder if AMZN will do (or already has done?) this and set the pace.

Thanks for writing and sharing this article, Jeff. It is great to be able to learn from your experience, and the candid and humble approach is appreciated!

Well, Unity was always easier to use. They focused on slick editors and tools early on.

It used to be that Torque had higher-end features, better performance and source access.. while Unity lacked many graphics and other high end features (they still don't have a great networking solution AFAIK), no source access but was way, way easier to pick up and use out of the box.

Now, with all the investment Unity has put in, I think it has most or all the capabilities you could want... and lots of other great stuff like first class cross platform support, including mobile. And it performs pretty well, at least as well as Torque by now, I imagine. Still no great networking, and still not even available source unless you pay huge (six figure+) fees.

Hard to argue that Torque should be the choice, at this point. Unless you want source access or a free solution. Unity has evolved into a pretty robust engine over the last 7 years, and its out of box experience and ease of use have always been pretty amazing (and fun!).

That's great. The Torque engine code improved tremendously over the years. In the early days, the code was rough, as it was inherited from a game studio that built it in a rush to ship a game and carried over from ~2001. It got better and better over time though and by 2006-2008, it was solid. Great performance, good features. Lots of the interdependencies at the system level were refactored and lots of the ugly bits of code (long inheritance chains, massive classes, etc) had been removed or deprecated.

Not sure how it continued to evolve later, or if they ever got rid of that scripting language... but it was on a good trajectory 5 years ago.

Glad to see it go open source, in any case. Lots of good memories with this engine and the smart, indie game community around it.

If you guys looked at several years ago, you should probably say "was" awful, as opposed to "is".

I haven't followed for years... but even circa 2007 and 2008 it had undergone huge rewrites from the old 90s and early 2000s nastiness with their simulation layer, scripting hooks, and high level classes. Was moving to a much cleaner componentized system with much better patterns throughout.

Or, maybe the codebase has actually gotten worse since 2006+, when it was shaping up pretty nicely. That could well be, I don't know. But such comments would be higher value if you at least reference approximate timeframe, as this product has been around in various forms for 10+ years, if I recall correctly.

As of a few years ago, the content pipeline had gotten quite good. I mean, not like Unreal or even Unity... but nice, and miles ahead of where Torque was in the early days.

To me, the big problems with Torque outside pipeline were: 1) Tooling (only the 2D engine and old 3d terrain builder had really nice WYSIWG tools 2) The damn scripting language... wish they'd moved to first class support for external bindings or an internal framework with standard 3rd party languages 3) Super old school, nasty game class structure (eg their "Player" class)... tons of inheritence and megaclasses that are just gross.

I think they were working on lots of that stuff before they sold the company, in addition to modernizing the graphics, etc. Don't know how far any of it got, but hopefully it is cleaner now.

Remember the engine was built in the late 90s and wasn't rearchitected for years, so it was what it was. Might be way better now, I don't know.

Well, that is good to know! Hopefully you are updating the site.

Can you provide any more detail on how it happened? Did a cousin's friend come over or something? Perhaps there are lessons for others to learn from this regarding securing your computer for guest use (ie I wonder if Sarah had access to your mail temporarily).

Congrats to Tog+Porter for their write-up this weekend in SF Gate: http://www.sfgate.com/style/stylebytes/article/Tog-Porter-ma... Looks cool.

This is gross.

Not the fact that the conference call was a mess of technical difficulties, or that Ed's boss interrupted his workflow.

That Ed sat back. That the company's culture would even tolerate anything near this level of lackadaisical disengagement.

If you're a Hacker, you don't say things like "that's not what I'm paid to do." You don't even think things like that.

Your job is to help make your company, and those around you, successful.

If you're not in a company with a mission you believe in, where you are excited to help achieve the goals the whole team is working toward... then freaking leave. Can you not find a job programming at a company that you do believe in? Then you're not a Hacker. You should probably stop reading Hacker News.

The failure in management here isn't interrupting Ed. It's allowing such a lazy, selfish culture to exist in the first place. It's not inspiring Ed enough that he actually wants to help his co-workers be more efficient, because he realizes they are valuable components of the engine that is trying to accomplish a shared mission and set of goals too. It's not firing Ed's ass any of the number of occasions where he has doubtless sat back and watched teammates make mistakes without even trying to help.

This is really, really gross.