a_d

Finland has 'baby boxes' delivered by the govt: http://www.bbc.com/news/magazine-22751415

Sweden has a govt sponsored program for end-to-end support for new parents: https://sweden.se/society/10-things-that-make-sweden-family-...

The reason for this comment is that HN has a 'startup' bent to it, so I hope some entrepreneur just Warby-Parkers (is it a verb?) the baby birth process. There is definitely room for massive simplification. Make each decision simple -- that alone is a giant opportunity.

Genuine question: What applications/problems it is useful for? Have you seen a good application yet (other than currencies)?

Would love to learn about the positives, from this community.

1. Lots of ICOs happening (crossed $3B in money raised).

2. ICOs are risky.

3. SEC will likely use the Howey test to determine which ones are securities (and therefore need to comply with securities law)

4. Startups may be able to protect themselves by limiting token sales to accredited investors and providing more disclosures in their white papers. (a hack that most people now are well aware of -- i.e. the utility token + SAFT route)

5. When you invest in an ICO, you not just take financial risk, but also may inadvertently be breaking a law (basically, ICOs are fraught with uncertainty)

I wish we heard something new. Clearly, there is something interesting happening - a new way to raise money, or perhaps a new way to launch products. In the long run, it is exciting because it will help launch useful products and services. But many articles often just regurgitate what most people already know, and don't offer anything new.

I suppose regulatory authorities at some point will offer a way forward (by regulating this space -- which I think would be a good thing!). Until that happens, we just hear a lot of armchair thinkers and potential practitioners just conjecture endlessly.

(Roughly speaking: IPython + Gdrive = Colaboratory)

Persuasion is the center-piece of business activity.

One of the most powerful ways to persuade is through storytelling. Well storytelling is central to our social and cultural experience. No wonder that it finds its way in our work — esp work at big companies (more humans = more social, cultural). Most presentations, discussions or meetings are exercises in persuasion via storytelling.

For years stories are told with theatrics or embellishment. That’s PowerPoint.

In the hands of a good storyteller — someone who has something to say and a compelling way to say it — PowerPoint will seem like a great tool. Conversely, the tool will suck (in the same way a guitar sucks when i play it, but becomes magical in the hands of jimmy page :))

Now, as a storytelling tool — just the fact that PowerPoint has found its way into a Billion computers — is astounding. Probably makes it one of the most useful tools invented!

There is a chance that all factory work becomes like a video game.

Tele-ops robotics would lead to:

- large-scale redistribution of physical/blue-collar jobs (bigger than "software" over last 20 years)

- allow new business models that were never possible in manufacturing (e.g. robots for "rent" - subscription model, gamification, flex-work like Uber for tele-operated robots in manufacturing; low-volume sales)

- order of magnitude better physical health of workers (lower injuries, more fun)

- savings for manufacturers (not just on labor, but lower insurance etc.)

- a 20 to 30 year hedge against the invention of general AI that seems far away and what other roboticists are chasing.

There is a potential to release huge latent wealth by creating a new industry (potentially trillions in value could be released over the next 10 years - no single company would capture it, but it would create something at least as big as "software").

Tele-ops is already successful for surgeries and dangerous situations. But it should be it’s own “field”

I think many of us have deployed this graph in our own lives. I watch movies this way - watch more movies made by a director I like. Then watch movies by the people this director likes. Its a wonderful exploration.

I wish there was a version of IMDB that allowed this meta exploration of movies - i.e. exploring the influence-graph for movies (or any art as a matter of fact).

There is a wonderful documentary called Connections [1] that explores this in the world of science, technology and engineering -- it traces the sequence of inventions and inventors that had influenced each other. I strongly recommend viewing it. You may find it very inspiring!

[1] https://en.wikipedia.org/wiki/Connections_(TV_series)

My guess is TransferWise captured the UK<> Poland corridor and expanded from there.

This problem is a great one to solve because the customers have a real pain and one can be very targeted about capturing high-value corridors.

I wish TW the best. Lower fee, Lower friction, Lesser time == Great for customers!

Anyone looking to build a startup in this domain should start here: http://www.worldbank.org/en/topic/migrationremittancesdiaspo...

(This data is useful to understand origin<>destination of remittance flows; Its a greater than $150B market and a lot of untapped/Uncaptured areas; lots still left to do!)

2. Instead of looking for problems, I think more about how is the world changing and what can I do about it. I would advise you to do the same. Sometimes something big changes and most people aren't paying attention. For example, Trump's recent XO on healthcare allows businesses to form "associations" (small companies can pool together to get large company rates). In the recent Berkshire meeting, Buffett said that American businesses are going to be increasingly uncompetitive (at the global stage) if healthcare costs keep rising. These are things that are connected - if one is paying attention. The uber thought is from Peter Thiel [1]

3. Instead of thinking about a startup, start an industry [2]. I think Alan Kay's talks are inspiring - but also give a good framework to think about projects.

4. If I could just give you one advice, it would be: Pay attention. Very few people are paying attention to the whats happening.

5. One of the biggest disadvantages that you have is that you already have a team and money. This puts enormous pressure to do something "Big". No good idea is going to survive this test - obvious "Big" ideas are taken (not that you still can't work on them). So when PG says, works on "projects" and don't think about starting companies - he means, just do stuff that you find fun. This is very hard to do in your case.

I have helped several companies when they found themselves in the same predicament as you did. If you think it would be helpful, I would be happy to talk: ad@42.life

Best wishes and Good luck!

To end with a cliche: Stay hungry, Stay foolish

[1] Peter Thiel: pg 114 http://gsl.mit.edu/media/programs/south-africa-summer-2015/m...

[2] Alan Kay at 1:10: https://www.startupschool.org/videos/11

"The way to get startup ideas is not to try to think of startup ideas. It's to look for problems."

I personally think this is an profoundly important piece of writing.

Ironically, the first "ICO" to do it with full compliance, useful tech and good business model, will be huge. There is demand for it.

The other big opportunity is to come up with a framework that everyone can use to be compliant (e.g. An Ethereum with "compliance" mechanisms built into it will be a big "general-purpose" blockchain).

It's hard to understand why ignore compliance, when being compliant is quite simply going to be much bigger, sustainable and "good business" than all scammers put together.

1. Fear of not communicating "everything" that you do. The fear is of being perceived as a very narrow solution when it does a lot more.

2. Advice that says "communicate the benefits" not "what you do". This advice could manifest itself in the wrong kind of (flowery) language. So instead of saying "export payroll reports for QuickBooks automatically" websites say "free up for time" [made up example]

3. Internal decision-making by committee.

4. Copying some website that you like - instead of thinking and reasoning from ground-up about 'what is it that I really want to say'.

5. Pretending to be a big/legit company when you are small

6. Big company with so many features that it would rather just show you the entire sales deck - the website is just there because it needs to be, but plays a tiny role in conversions. [why focus on something that doesn't add value - in this case, the website e.g. SAP.com]

This is a good article. Everyone who is running an online would benefit from thinking hard about this.

1) Protocol coins (coins that have some value within the project / product) and

2) Asset coins (coins that can be thought of as shares in a company).

Protocol coins are being positioned as presold revenue and need not necessarily conform with securities laws. Most ICOs are trying to position their coins as kickstarter-type "token gifts" (e.g. T-shirts)

Asset coins on the other hand, come under the purview of securities laws and act

People are buying these coins because of FOMO. Large majority of these projects having ICOs are scams - not too different from the dot-com bubble. Eventually, some good ideas are going to come out of these projects.

(I have no association with Floyd)

I like this video because of the emphasis on the role our psychology plays in the search for product-market fit. Understanding people/users well - what they mean when they say "this is interesting" is really important. Systematic de-biasing of our own mind is an important idea in this talk.

Richard Feynman's famous conclusion to his report on the shuttle Challenger accident said "Nature cannot be fooled"-similarly, markets/users cannot be fooled. Either a product or service solves their problem or it doesn't. Anything in between needs to questioned carefully and observed (in an unbiased way) closely. Feynman also says "The first principle is that you must not fool yourself – and you are the easiest person to fool". Same idea as Peter's -- and a very important one!

I wish more founders talked about "what product-market fit feels like" (once they found it) - because only with enough samples, new startup founders can better understand whether they have achieved pm-fit. However, I wish video shouldn't be mistaken for thinking: "my Show HN post tanked, so I don't have pm-fit". I think it is worthwhile trying to understand what he means. It is quite clear when you really solve a problem that someone else has. It is also a great feeling.

One of the reasons why finding pm-fit is hard is because there are too many variables. Peter simplifies it to some extent when he says that while searching, one must be clear if you are changing the market or the product. This is correct, but in reality, these adjustments are quite subtle. So paying close attention to these constant/rapid adjustments (stepping back and thinking about them) is quite important.

I like the conclusion of his talk: you will know when you find it.

[1] http://pmarchive.com/guide_to_startups_part4.html

Just learning how to build tunnels is interesting from that respect.

The reason that isn't stated is because Musk likes to create projects around a "purpose" and business plan for them to succeed (even if they are retro-fitted - that is, he decided first that it would be important to tunnel, and then arrived at the "why tunnels on earth").

edit: not a new thought. some other commenters already pointed this out. I read those after posting this.

The post says that the imitation network takes the input from the vision network and processes it to infer the intent of the task. Isn't the "intent" always "to stack"? Or can the imitation also be just re-arranging blocks in another configuration?

This part is interesting, if I understood it well.. > "But how does the imitation network know how to generalize? The network learns this from the distribution of training examples. It is trained on dozens of different tasks with thousands of demonstrations for each task. Each training example is a pair of demonstrations that perform the same task. The network is given the entirety of the first demonstration and a single observation from the second demonstration. We then use supervised learning to predict what action the demonstrator took at that observation. In order to predict the action effectively, the robot must learn how to infer the relevant portion of the task from the first demonstration."

Does this mean that the imitation network has been trained on stacking, un-stacking, throwing...and other such tasks, and then it identifies that "stacking" is what is being done in order to imitate it?

Is there an ELI5 for what the 2 NNs are actually learning?

You are right about the philosophy. But a thought regarding your specific example: even if self-driving tech were to be available tomorrow, it will take many years for all the cars (the "fleet") to be self-driving. There are about 2B cars on the road, and every year about 100M new ones are made. It would take 20 years to replace the fleet. So the "parking problem" isn't made immediately redundant with the invention of self-driving tech. There is still room (and in my opinion) worthwhile for a startup to solve the parking problem (however they attempt to solve it). Just saying that I wouldn't discourage anyone from attempting this :)

1. Live in the future: If you live in the present your ideas are going to derived from the present, and therefore will be incremental. Paul graham points to a similar idea in his essay here http://paulgraham.com/startupideas.html

2. Learn the rules and break them like pro: An important idea that says learn deeply about a field (industry or language or medium - whatever) before you go about breaking its rules. This is something that should be discussed more. On one hand, this implies that rules need to be broken carefully by understanding something very well. On the other hand, a certain "naivete" also has been proven to be successful (As he himself says in another portion of this talk). Requires more thinking... I have to say that I have worked with two geniuses who changed industries (caused a "Disruption"), but while they were "outsiders" they understood the industry they were changing very deeply, and could point to very specific flaws in them (legacy rules) that were ready to be broken.

3. Scientists and engineers working together: I love this specific idea. He points to the Manhattan project and the Radar project as successful examples (and ofcourse, Xerox PARC itself). I also understand why OpenAI is organized the way it is -- they must have taken inspiration from Alan's ideas.

4. Progress is the only important thing: This is very similar to YC's idea of have a metric and grow it by X% week over week. One of the big things (IMHO) that YC cracked was to find way to fund hard tech problems, using this very specific idea that worked very well in software. Break a big problem into small parts and show progress on it every week/month/quarter etc. It is a simple and elegant way to think (and work).

5. Advance something very important: This idea is dedicating your life to an important goal / problem. This idea stirred me the most (for personal reasons). This especially rings true for me, esp when I meet someon who found "overnight success" after toiling aways at something for 10-15 years. This idea requires a lot of reflection. Alan says: "Fix big human problems"

6. Argue for clarity, not to win: An amazingly simple yet powerful message. So much of our interactions would be better if we (if I) did more of this.

3 Sidenotes:

a) I wish he had answered the question about "What caused the regression?" that someone asks (I think Sam) after watching a demo of an early "iPad" from the 1960s. It is really important to understand this. Elon Musk often says that tech advancements don't just happen automatically, some group of smart people need to work together to make it happen. I think something happens that governments, VCs, technologists stop working on advancing tech in some areas. Some of it is explained by lack of a "great adversary" (WW II, Cold War etc), but still it would be great to understand this very deeply. Thiel's famous book also talks about regression in tech, without fully explaining it. It is an important observation, that merits a deep discussion. I would be curious to hear what Alan says. Maybe @Sama can fill in (?)]

b) Special mentions to two ideas, not because of the ideas, but because how he puts them. a) "Fund people not projects" (basically how early stage funding works) - but Alan says, these are "artists are people who do their art because they must". And b) While talking about "build your own tools if you have the chops" he says "...Otherwise you are working the past on some vendor's bad idea of what computing is about". I love how he puts these thoughts across.

c) I recommend the video @_pius links to in the comments: it is called "the power of simplicity". One of the great Alan Kay talks.

And Q2: when you say "took off on its own" does it mean folks learned about WhatsApp only from word of mouth with no effort from the WhatsApp team?

Q3: what impact did charging for the app have on the company finances and growth rate?