plam

dupe https://news.ycombinator.com/item?id=29682051

dupe https://news.ycombinator.com/item?id=29682051

if you're interested in this, you might find this alternative interface useful -- https://hckrnews.com It keeps track of top links each day and you can scroll back in time. This has been helping me check HN less frequently as I don't feel like I'll miss out.

It seems one of the biggest business issues is site visitor perception. I have a couple questions please.

What's your opt-rate rate so far out of the sites that have shown this option?

do you have any real data regarding hidden cost to a site visitor? like power consumption A/B testing or something?

email sent

That's a very real problem I totally forgot. We could do the same thing and have a small team of interns + mentor do maintenance mode work.... but not exactly exciting. I don't know. Any idea?

You can buy all sorts of Chinese phones (with dual SIM) directly from China online on sites like Aliexpress.com. The thing is, US uses weird 3G frequencies or something from the rest of the world.

I bought my 3G MTK phone a year ago on Aliexpress.com. It worked fine perfectly while I was living in Europe. Then I came over and now I'm always on Edge network and network speed is significantly slower.

do you have a few minute to chat in the next few weeks? drop me an email please paul@quantisan.com

yes, a large part of it is figuring out the problems and where are the lower hanging fruits and what resources are needed.

drop me an email please paul@quantisan.com

I'd love to hear of your experience.

lots of great suggestions here! I'll reach out to local schools, bootcamps, etc. to get some feedback

what student would get is a tech mentor and a chance to solve real problem with a good cause. wouldn't that be a competitive selling point vs. a paid internship with perhaps a few thousand dollars? Perhaps I'm being naive / optimistic?

good question, I edited post to clarify.

not yet, but probably eventually. I'll start with just doing it for a project first to test out the model. I have a nonprofit on board with needs. I have connections with developers. I can find an intern. I can get most of it in place but lack time and experience (haven't done this before). So I want your help to iron out the model as much as possible to not waste time.

that depends. to pay students I would need to source funding too. that is doable but does add another level of barrier to get things going. I'll probably start with finding unpaid to get a few batch going, gain momentum, then if all is well, we can start looking for funding.

I did a piece on using Clojure with Go recently. For me, it wasn't a switch but using them along side each other to tackle different problems more efficiently. Here's the link -- http://www.quantisan.com/simple-easy-quick-using-go-along-wi...

Much of Clojure itself is written in Clojure. I would recommend going right to the source for idiomatic examples. https://github.com/clojure/clojure

I wrote some R scripts to perform historical data-driven asset allocation for long-only, low-load funds investment for the long term. This seem like a good time to put it here: https://github.com/Quantisan/touzi

It scrapes not only price data but management fees, etc to minimise cost and risk. The goal is not to make profit but just to keep up with the market but at a lower risk.

Regarding ground changes, we can look to current bullet-train rail systems. Nobody has ever died on a bullet train in Japan due to earthquakes. I saw on a documentary that Japan feed seismic sensor data into train control centers where any dangerous seismic event would trigger automatic stopping of the trains.

I share your disagreement with the piece. Not everyone needs or wants to be "fixed". It took me years to realise that the best I can do for others is not to listen for a solution to "help" them, but listen just to listen.

SEEKING WORK - Remote for companies on US times; local in London negotiable

Data consulting to make the most out of your data.

I am a full-time data scientist actively involved in the big data scene in London and is a core committer on an open source MapReduce tool called Cascalog written in Clojure.

I have delivered projects such as i) extracting features from our customer event data on Hadoop into Neo4J to drive a graph-based recommendation system to increase cross-selling, and ii) replaced our SQL servers data warehouse to one on Hadoop and Amazon Redshift to make messy data usable for the rest of the company.

See my blog and talks at http://www.quantisan.com

If you want to figure out what you can do with your data to grow your business, drop me an email at paul@quantisan.com

my whole office use emacs for clojure dev but I use vim. I tried emacs for a few days on more than one occasions but simply couldn't bear the physical awkwardness of the key combos (I did remap ctrl to caps).

do they?

I apologise for my naive misunderstanding of the Analytics metric. I've updated the post to point this out at the top of my post. Thanks for the explanation!

I assumed they have some fancy javascript thing to take care of that? what would you suggest to use to estimate actually readership?

good point, I'll keep an eye on returning visitor metric over the next few days

at this rate, I'm now hoping that I can do an analysis of the analysis of a viral post that also have gone viral. how awesome would that be? :)

thank you for understanding and clarifying my intent

I never wore mine. Got one mostly to be able to go through the ritual with friends.

OT: Funny to see that the term 'engineer' is held in high regard in Canada but here in UK there's no such connotation.

Correct me if I'm wrong, conceptually, probabilistic programming does to equation solvers what declarative language like SQL do to query data.

Say you have a model, y = mx + b, and you have some data. Normally you can solve this explicitly with matrices or whatever. But you'll need to implement the solution. With probabilistic language like BUGS, you can feed in the model and the data and it would return `m` and `b` as a probability distribution.

This is an trivial example. One example of where it really comes in handy is doing bayesian inference.