scikit learn is a great package, but it's bizarre that people haven't heard of it before?
660 karma · joined December 13, 2013
One problem I have noticed is innapropriate downvoting. For example, if a voter downvotes a comment simply because he/she disagrees with the opinion of a commenter. How do we encourage a diversity of opinion on a site of like-minded people?
And similarly, how do we prevent "downvote conformity" (I apologize for my invented phraseology, but I'm creating a name for the concept that our community upvotes posts that are generally agreeable and downvotes posts that may be less mainstream, thus imposing a conformity of opinion?)
While they could definitely make all of their events streamable on a wide variety of devices, this tactic makes everyone not in the ecosystem feel as though they've been let down by the technology which they own.
It also ensures that the audience is more Apple-centric, and the first to hear the conference lectures will already be biased in favor of Apple given that they are already Apple users.
Yeah, I guess you're right that thinking is too hard and we should simply listen to a population with a sample size of 1.
1. Searches google for a restaurant name 2. Checks the number of stars Yelp gives it in the Google result
or
1. Searches "dinner" on yelp 2. Clicks on the highest rated Yelp pages
For a popular restaurant with a fan following the Botto pizza strategy might work. Generally speaking, ignoring your customer's needs and wants is a fast path to complacency. And hopefully you're all aware that:
COMPLACENCY WILL BE THE ARCHITECTURE OF YOUR DOWNFALL.
Internalize that statement. While Botto's is cool today, adaptation is the key to long term success.
Inevitably after serving each of their loyal customers many times over, the customers will have a bad meal here or there. With an attitude of "The customer is not always right", disagreements will leave customers disenchanted one by one, and slowly these loyal customers will become less loyal. Or another good Italian pizza place will open up. Or the italian food wants of the neighborhood will change. Or Yelp's powerhold on connecting diners to restaurants will grow stronger.
But I do think hackernews should consider making some articles locality specific. For example, a user logged in from Berlin could be shown an article of geographical interest. You can take this a step further and hsow articles to a user based on their interests. For example, a developer might be more interested in coding articles and a product managaer or vc might be more interested in new products.
It helps when people get personalized content, because a greater diversity of articles might get read by the community, and then truly interesting ones will become front page.
1. For KPCB to exit, they need to find an investor that values the company more than they do. To find an investor who values an unprofitable picture sharing company at over 10B would be a difficult task. It will probably be possible to find retail investors to provide the exit strategy, but the big tech companies of the world (Google, Facebook, etc.) do not have incentives for buying out unprofitable poor performing companies with no future. They have an agenda, and their CEO's are looking for strong future revenues because they themselves are heavily vested in the companies which they work for. These are savvy people with a team dedicated to valuing M&A transactions. There is no good reason for a large tech company to acquire a snapchat unless it thinks it will contribute to the bottom line, the brand, the network, the IP, or the scale of the existing business.
2. There's this incorrect notion that VC firms are somehow not exposed to the risks of the market. Times are good right now, and if you're a VC firm, you should be making a profit right now. A typical pump and dump occurs because of timing effects, insider knowledge, and mass marketing. While you could argue that an IPO is a "mass marketing", I think that external financial institutions (hedge funds) are very much looking out for shorting opportunities. An overpriced IPO is a great opportunity for a hedge fund. Also, It's worth mentioning that the VC firms which will profit so nicely this year, are likely to lose 50-80% during a downturn. They are very much exposed.
3. If KPCB really wanted to profit, why only invest 20MM? This won't make them much money. The real reason they made this investment is because they're skeptical of snapchat, but still want the talking point if there is an upside. It will be a great story for investors, pretty much regardless of how snapchat exits.
Imagine a world where daily time-series data can be stored efficiently: This is a lesser known use case, but it works like this: I'm a financial company and I want to store 1000 metrics about a potential customer. Maybe the number of transactions in the past year, the number of defaults, the number of credit cards, etc.
Normally I would have to duplicate this row in the database every day/week/month/year for every potential customer. With some kind of git-like storing of diffs between the row today and the row yesterday, I could easily have access to time series information without duplicating unchanged information. This would accomplish MASSIVE storage savings.
FWIW efficiently storing time series data is big problem at my company. No off the shelf solution makes this easy for us right now, and we would rather throw cheap hard disk at the problem rather than expensive engineers.
I think it's crazy how fat fingered people have become with the down vote trigger finger. I think it's 100% crazy to discus the timing attack without considering how it could be thwarted.
I understand that it's tough not speaking English as a native language. I understand that you don't feel like you have a lot going for you socially. At the end of the day any one person can come up with a million reasons why they're not "cool".
I program all day. I have bad breath. I don't drink. I have trouble hearing. I have a soft voice. I get really sleepy.
It up to you what you want to let hinder your social life. In high school, I was extremely introverted and had little experience talking to girls. To build up my confidence, I started working out, reading magazines like men's health and GQ to build out a more stylish wardrobe and pick up on some social queues. And slowly I became friends with a few other people. We weren't the coolest bunch in town, but by the end of my senior year, I was certainly one of the best-known faces on my high school campus.
Of course, then I realized years later that while this was all good, I had lost sight of one of true passions: programming ;-)
HP has been working on the memristor since 2008. Memristors have already been produced in labs by U of Michigan: https://en.wikipedia.org/wiki/Memristor
This article seeks to give the image that this is a novel idea that investors can take advantage of. In reality, it is not.
I guess putting technology in the hands of the non-visionary executive with a strong business sense is basically asking for your [tech] company to die. You need engineers or at least visionary product people in management positions making these kinds of decisions. While it would have been sad if excite had bought Google, it would never happen because they didn't have the vision. And if they did get the vision, Google probably could have continued to become a semi-success in that environment (although probably nothing like what it is today).
Ideas are inherently not obvious. Judging the value of an idea is even less obvious.
At the end of the day, you need to take small bets. Don't wager your life savings. Listen to the criticism and have the confidence to filter out what you think is irrelevant.
"First they ignore you, then they laugh at you, then they fight you and then you win."
Think of it like this: it's really easy to push a car that weighs one pound. Much harder to push a car that weighs 1000 pounds.
The momentum advantage was explained in the first point.
You have one car that's 1,000 lbs and one that's 10,000 lbs. When you start each car, it will take a lot more energy to bring the 10k lb car up to the same speed as the 1k lb car, but after that, the effects of drag will weigh more heavily on the lighter car (F=ma, so the rate of deacceleration is a lot higher for a smaller mass), causing constant fluctuations in the motor shutting on and off.
With the the 10k car, the transitions would take much longer. The time in between accelerations would be longer, and the motor would be on for longer spurts. You gain a great deal of efficiency for simply leaving a mechanical system running vs constantly modulating a mechanical system (i.e. shutting it on and off).
This would work in favor of the heavy car being more efficient.
Of course, the heavier your car is the worse your 0-60 split would be, so for a performance car that's a negative.
Also, the heavier car would experience greater bearing resistance + friction. This is really difficult to quantify, and my gut is that it's outweighed by the momentum advantage.
But overall, for a car that ways 2k lbs, a 5% increase in mass is pretty negligible, especially for casual highway and local driving.
While it requires more initial force, that extra energy due to added mass is stored in kinetic energy, and contrary to popular belief, it does not affect efficiency. There was actually a myth busters episode about this.
Electric cars would store this added energy in their regenerative breaking, so it shouldn't affect energy consumption.
Also, the added weight of steel over titanium would be negligible, maybe 100 lbs tops.
Titanium has unbelievable tensile strength for its weight, but there's no good reason to make an "underbody shield" out of titanium except for publicity. It would make way more sense to use steel (and maybe you could make a case for something ultra light weight, like carbon fiber, but probably not).