jw2013

There is a counterpart company in China called FlashEx that is doing pretty well (just raised Series D in 2018) right now, focusing on same city on demand shipping.

shoot me an email if you wanna chat: (dedacted since we have chatted) :)

Looks pretty legit. Keep it up!

For the language I love its concurrency model; creating thousands of "processes" at your PC/laptop is no problem. Message passing over shared memory also makes your distributed program easy to reason about. When talking about Erlang, you can't avoid OTP. It makes creating fault tolerant system at ease. Putting these two pieces together you have a very scalable fault tolerant distributed system.

I don't like the syntax of erlang though. And the missing of macro makes me switch to Elixir.

For one thing, how many drivers they have matters. More data might produce better estimated ETA (at least Uber use the data from the drivers to predict future traffic congestion, etc.).

Another thing is how well they control supply and demand. Say Uber has 10x drivers than Lyft, what if Uber has 100x consumers than Lyft? Uber likely has to surge at a very high rate, and the consumer can't afford a surge might go to Lyft instead. Since Uber can't control well how many total consumers they have at the moment without surging, one realistic alternative is to just reduce the number of Lyft drivers, so Lyft have to charge higher rates as well.

Here Uber doesn't just want more consumers coming in, because even less consumers is better than more consumers come in initially and can't afford a surge and then have to turn to Lyft.

So to sum up, in the end, Uber have to recruit more drivers (hey, they even just opened API for their product) or cut the number of drivers from Lyft (through alluring them). You can't blame Uber on that, it's the competitiveness of this market.

Did it ever happened any investor to YC startups tried to take advantage of the founders before, or this post is just an act of prevention?

As the ban of ZenMagnets's primary opponent was in 2012 [1], and the government seemed to want to ban ZenMagnets [2], I find it amazing it is still out here in the business two years later and with a record sale $700k last year (a jump from $50k in year 2012).

So nothing happened during these two years?

----

[1] Banning Buckyball Magnets Is Statistically Ridiculous: http://www.huffingtonpost.com/eliyahu-federman/banning-bucky...

[2] US Government Wants to Ban Zen Magnets: http://gizmodo.com/5932638/us-government-wants-to-ban-zen-ma...

You can always talk about edge cases, which do not reflect the reality for vast majority of women.

At least the window to give a birth to a child for a average woman is a lot smaller than a average man.

The context for these coding-challenges is they are the type for algorithm competition or white-board coding interview.

Back then when I was still looking for job, I was asked this question by a startup. I gave out this hashset solution and was quickly asked if O(1) space solution was available and then asked to implemented it.

If the space restriction is not an issue, I would definitely go with the method you suggested. Way more succinct and easier to follow.

The function you wrote:

    a) takes O(n*lgn) time; the method in post uses O(n) time
    b) use extra memory; the method in post uses O(1) extra memory
    c) have logical error: the problem asks to find the first missing positive (in range [1, infinity)). 

For example:

    a) your firstMissingPositive([100]) gives ValueError: min() arg is an empty sequence
    b) your firstMissingPositive([]) gives IndexError: list index out of range

This is all in Jeff Knupp's Writing Idiomatic Python:

http://www.jeffknupp.com/blog/2012/10/04/writing-idiomatic-p...

Many so-called Python Idioms are really non-intuitive and I don't really appreciate.

If these people are first-time founders, what (characteristics) will you or other good angel investors looking for in them?

Is it convenient to ask if you invest in some decent proportion of first-time founders?

While I do agree there should be less "activation energy" for the newbies to enter the Erlang community (https://www.youtube.com/watch?v=Djv4C9H9yz4), I don't think "identify ways to expand the scope of Erlang and Elixir to other potential use cases" is the way to go. Each language has its strength and weakness, being highly concurrent and fault-concurrent is the design goal of erlang/elixir, and erlang/elixir achieves that. Making them a general-purpose language seems a bad idea to me (think message-passing a large size of message in a non-concurrent situation where the shared-memory should be used. it just deteriorate the performance of the general-purpose program). Being able to built a great concurrent system is enough to make a language become popular in industry, the problem is really not about more use cases, but is how to make the barrier of entry lower. Also, it does matter to have a warming community, like most of Ruby communities I know.

No way they are taking the traditional 7% ~ $120k deal. From Crunchbase they already got $141M fund, and let us be very conservative and say their valuation is just $400M. 7% percent of that is 400,000k*0.07 = 28,000k.

I guess Quora will be treated the same as other batch's participants (except I wonder how many of Quora's employee will be physically in SV), but Quora got different deal.

by 2020, the biggest tech company in the world — Apple — will cease to be the most important, and won’t even be in the top three....he predicted that the top three tech companies, instead, will be Google, Facebook “and one that we’ve never heard of.

Okay, a couple of questions here:

1. Is Apple the top three most important company even right now? Don't mistaken me, I love Apple.

2. Like many doubters here, I wonder if Facebook can even be relevant in 2020. Google succeed because it does not limit itself to its original service (searching) and created tons of other cool services that a lot of people want. The question is whether Facebook can pull that off? Not yet at least.

3. If there is some company beating Google I am thinking of a company in AI with some tech breakthrough. But then the question is, would Google have enough smart people to "copy" it and make it better? I am somewhat awed at what Google can achieve at this moment. Next Google please comes up.

Yet another respectable move for YC. Hopefully this will spur more global interest in starting a startup. One thing I really appreciate YC in recent year is making the startup school videos online so those can't attend will not miss the talk. Of course those attend will meet tons of amazing people. Bringing in smart and talented people into one room just sparks insightful conversations. First in valley, and now globally.

Thanks, YC. That is so unselfish.

Great article.

> If you join a company, my general advice is to join a company on a breakout trajectory. There are a usually a handful of these at a time, and they are usually identifiable to a smart young person... Spending a few years at a company that fails has path consequences, and working at an already-massively-successful company means you will learn much less, and probably work with less impressive people.

I actually find it opposite. First of all, it is not so easy to identify a company on a breakout trajectory. The line between a breakout success and failure is just too thin on an early-stage startups. Even if the startup works on the same field as your expertise, you have been following that field for years, and you think the founders are brilliant, you can still get the judgement wrong a lot of the time.

I think a startup's failure is not so bad for a young and ambitious employee. There are just as many lessons can be learnt from failure as from success. As for path consequences, the young kids got so many productive years left and one failure is not really destructive or harmful. And a great thing is, if a startup fails most likely it fails fast. For your next job, people won't hire/reject you because the previous company you worked for failed, they hire/reject because you are brilliant or not. Likely the stuff you learnt in the failed startup will make you a brilliant person (tech and non-tech wise).

No offense to the author, but what a terribly written article full of logic holes. For example:

'Our parent company's former CEO started off as a programmer 25 years ago, switched to manager; in 15 years he went all the way to being CEO of a $4B company. After 10 years he retired recently with mansions and cars and no worries. Meanwhile I work with people who started around the same time and who are still senior software engineers.'

So the author just picks one person that happens to be the CEO evetually? What about the rest majority of managerial forks? They probably are stuck in corporate ladder and even may be fired already. The chance of a person working on a corporate managerial job evetually becomes a CEO is likely no higher than a tech person eventually becomes a CTO. So why just pick one CEO person to illustrate managerial role is more promosing than tech role? Comparing the max value of a group to the average of another group is unfair.

Also, I program because I love programming. Even if you gave me one billion dollar making me manage people for the rest of my life instead of writing code, I would still turn it down. If doing the thing I love is dead end, then I beg a different definition of dead end. Okay, just use author's definition of dead end as slight chance of moving up, but I just don't care as long as I love the job I am doing.

__

An interesting side note is programming as a hobby. But I would still rather programming both at work and a spare time, as long as the day job I do making me happy.

This basically reduces to two questions:

a) is github monoculture is bad?

I really disagree with author's view on the emphasize on public github project in job interview process is misplaced. Yes, it will not eval how good you are at your previous job, but the problem is it is very very hard to find how good you are at previous job. Doing open source project at spare time is a positive sign of interest in programming. If in the interview process you said I don't have a github account, but I put all my repo at abc.com, I am pretty sure it will work, too.

And I don't care the monoculture if its service is the best in this field. Why use the second best if the best is possble? I would like to be an earlier adoptor if there are better service, even if there are currently no user.

b) how to disrupt github?

That I have no idea. What github does not provide but is really needed? I do not know. Would love to hear ideas.

The primary reason I won't work for Google is because I'm nowhere near smart enough.

But as long as the hiring process keeps the same as of it is now, it can be hacked not that difficultly by keep practicing on coding exercises (e.g. http://oj.leetcode.com/). All of my friends spent several weeks working on the coding exercises got into Google (for those cared to apply). Several of them are really not that "smart" and they are below-average in project-coding & general computer science understanding (sorry my friends), but repeatedly working on coding exercises and even memorize some common problems' solution just works for Google or Facebook's hiring process.

That process of cracking the coding exercises can be boring, though. I would be immediately returning to do my side-project even after I do one or two coding exercise. Not worth of the time at least for me; there are tons of great companies packed with people smarter than the average Googlers and they don't emphasize on algorithm in hiring process that much, why not work for them?

I once wrote a kernel loadable module for getting page fault handling info (time cost, etc.):

https://github.com/jw2013/getvminfo

I use that to see the page fault pattern of both sequential and random memory access pattern. In case anyone is interested, just check it out.

I think the author fails to realize whether saying 'easy' is okay or not really depends on what background your audiences have. For example, in the math class I find the professor saying 'it is trivial to get B from A' is quite okay if the proof of that deduction is something the students have learnt in the lower-level classes. I don't know another way to skip the easy proof without saying it is easy.

If you know how well the person you are talking to know the material you are talking about, then I don't find calling the thing 'easy' is humiliating at all, instead it is a effective way to say "no worries, just do XXX and you are good". A rule of thumb is when you are not sure if it is appropriate to call something easy, just ask the person you are talking to whether he knows that thing. Why stopping using the word 'easy' when we are sure all parties of the conversation gets what you are talking about and think it is indeed easy?

Someone here mentioned Amazon's core value on frugality:

"We try not to spend money on things that don’t matter to customers. Frugality breeds resourcefulness, self-sufficiency, and invention. There are no extra points for headcount, budget size, or fixed expense."

But isn't making the employees happy directly and indirectly influencing the product and service the customer is getting at? I understand the business Amazon is in is of low-margin profit. But still, high employee turnovers may cost more than the extra-money the company can spend on making the employees happy. Not to mention the employees productivity can increase when they are happier, so the ratio of (productivity/salary) still increases even though salary increases. May be now Amazon can have less number of employees, but each doing more productive work because they are happy.

gittribute (http://gittribute.herokuapp.com/)

Gittribute saves your time on getting help on your open-source project. Need help on your project? All you need is to add one line on your Github project's README.md & click your link there, and then everyone in the world will see you need help and they will find you.

It is under MIT license here: https://github.com/jw2013/gittribute

I want to post a screenshot, but currently there is only one project listed on the website, so may be it is better if those interested care to check the website itself: http://gittribute.herokuapp.com/

Even it is not explicitly mentioned, I wonder if Stripe will let 'whether the applying project can be used somewhere inside Stripe' be one deciding factor, or the project can be totally irrelavant to what Stripe is doing.

It is great seeing YC companies giving back to the community, first teespring and now stripe. I can see it is the future that it is startups pushing forward the technologies and communities, much more than big corps.

I doubt there will ever be any clear explanation on what happened. Two sides of the story basically tell two different stories. As someone mentioned in the post, Julie posted some tweets in response to Github's announcement today. (https://twitter.com/nrrrdcore)

1. Bullying someone into quitting: Illegal.

2. Asking an employee to relay private conversations between her and her partner: Illegal.

3. Justifying the harassment of an employee because of her personal relationships: Pathetic.

4. How does it feel to make money for liars and cowards?

5. Pushing women with strong opinions out of your company because they disagree with you is wrong.

6. What number am I on? Oh yea, how do you sleep at night?

7. Leaving GitHub was the best decision of my life.

8. There was no investigation.

9. There was a series of conversations with a "mediator" who sought to relieve GitHub of any legal responsibility.

10. Whose reasoning included "would it surprise you to hear that [your harasser] was well-liked?"

11. No, no it would not.

12. Women at GitHub who sprang forward to defend the men who harassed me, it is naive to think the same thing cannot and will not happen to you.

13. Best of luck rolling the dice.

14. A company can never own you. They can't tell you who to fuck or not fuck. And they can't take away your voice.

15. Unless you let them.

16. Hmmm still no mention of the man who bullied me out of our code base because I wouldn't fuck him. Too popular to be accountable, I guess.

I would love to contribute to Elixir, but jumping right into Elixir's issue list on Github to fix the bug there is a bit too hard for me. I wish there was some resources getting us familiarized with Elixir code base, and leading us more easily into contributing to Elixir, just like the The Eudyptula Challenge for Linux Kernel (http://eudyptula-challenge.org/). Or is there already this kind of resource?

Can anybody tell me what company he is starting? From the article it seems vague and probably he does not even have a specific idea yet. Would love to join if the plan is more specific.

So I took a quick search and find a list of 50 code bootcamps, and many are duplicate bootcamps on different locations:

http://www.programmingisnothard.com/bootcamps

A few things I noticed (through just a quick inspection):

  1) Average training time is about 10 weeks;
  2) 28 out of 50 teaches Ruby/Rails, 15 out of 50 teaches JavaScript, 5 teaches Python, 3 teaches iOS, and 3 teaches Android.

So, yeah, the curriculum is pretty practical. I doubt many of the curriculums cares about C.S. basics as long as it is enough to let them write Rails code.

I think one way to filter out the inept students is may be let them do some algorithm problems during interview? Also if your company does Ruby, then let a part of interview process be solving the bugs or issues on company's current project, and try to see how to approach the problem and whether can get it done. Understanding other people's code and reproducing/tracing/fixing bug is something very important. Also a big thing is when a bootcamp grad candidate say "you should do X in Rails", probably ask them why is a good idea. Do they do X just because their teachers say so, or there is a good reason for really doing so.

Disclaimer: don't have any experience with any bootcamp grad, but with experience working with X (place Rails, Django, etc. here) only guys.

But the problem the author described in the article can be solved just by better programming habit- always add braces even for a one line condition statement.

Or just put the one line statement in the same line with the conditional statement such as: if(condition) statement; so when you try to add a line next time, you will notice it was a one line if statement.

But yeah, not explicitly requiring braces for one line condition statement can give us more succinct code but does requires better programming pratice.