blahi

16 ms·

121.
by blahi 10y ago
But it was logical.
122.
by blahi 10y ago
It is also worth noting that Galileo had a very big mouth and had many enemies. The publishing of this work was a perfect opportunity for them to pounce. It wasn't so much the church hating science as much people hating Galileo and usi
123.
by blahi 10y ago
Genomics datasets can be several hundreds of petabytes. I've heard about an exabyte+ dataset too.
124.
by blahi 10y ago
Then why do you reply to comments directed towards somebody else and act like I was talking to you?
125.
by blahi 10y ago
You have 2 accounts commenting about the same thing? Not cool.
126.
by blahi 10y ago
It's totally the same. USPS should adopt flat structure because Valve does it.
127.
by blahi 10y ago
>But if it means there is a central intelligence (democraticly controlled, like by a parliament and not a dictatorship!) that can override individual business decisions in important corporations, then sure We have that. It's just co
128.
by blahi 10y ago
>I'll never understand the fascination that governments and companies have with top-down modeling of demand. Because that's the only way to manage an organization that consists of more than a dozen hackers. >You get gluts or
129.
by blahi 10y ago
wat? Azure has minute billing... So does google. AWS is a mess.
130.
by blahi 10y ago
You have $25/month for free for a year... And $200 free trial. You also get credits when buying VS/MSDN.
131.
by blahi 10y ago
>In some aspects they're playing catch up And it some aspects, they are way ahead. Analytics is one of those examples.
132.
by blahi 10y ago
>Microsoft taught R to interact directly with SQL Server's storage engine They have created 2 new services for SQL Server 2016 - BxlServer and SQL Satellite which facilitate the communication and data exchange. They obviously have a
133.
by blahi 10y ago
>being able to initialize and share data between functions within the R runtime That's right. >rather than having to transfer back and forth between Postgres and the runtime But that's not. There's a difference between
134.
by blahi 10y ago
If you are so sure of yourself, model it. http://www.lunametrics.com/blog/2016/04/21/compare-actual-vs... https://google.github.io/CausalImpact/CausalImpact.html Because some of the
135.
by blahi 10y ago
>The point of the service provided by Google is that it's on the whole, significantly easier than "traditional" ad buying What the hell gave you that impression? He does not owe any appologies. He is completely right. How
136.
by blahi 10y ago
Yeah, ask them how they felt about it when they turned adwords off.
137.
by blahi 10y ago
No. They have their own highly optimized algorithms. They also have their their own distributed data structures. I am not sure about that 2nd link. Seems like it's just UDFs written in R.
138.
by blahi 10y ago
No, it is especially appropriate given the audience here.
139.
by blahi 10y ago
You would be surprised how many people click on ads but say they don't. They just filter out that information and forget.
140.
by blahi 10y ago
I would also suggest The R Inferno.
141.
by blahi 10y ago
You might be facing very byzantine leaderships. But you might also be severely overestimating your skills. Not only your technical ones but your political ones. A key requirement for a leadership position is being convincing. Social intelli
142.
by blahi 10y ago
Out of core algorithms, attatchment of data straight into R for a couple of very important things.
143.
by blahi 10y ago
That's a lot less than what Microsoft is offering.
144.
by blahi 10y ago
How would you feel if somebody who saw what programmers do only on TV tells you "A very highly trained monkey can do what you do - type on a keyboard." Because that's what programmers do, isn't it?
145.
by blahi 10y ago
In that particular case, I used Vertica which loads data in R really, really fast and straight up use a very big machine. That's not how I approach it most of the time though. I mostly use out-of-memory algorithms, sometimes open sourc
146.
by blahi 10y ago
The growth during those 14 years was factored in the stock... This exact thing was actually discussed on Martin Shkreli's stream a couple of weeks ago. Maybe you shouldn't have such strong opinions about things you perhaps don&#x2
147.
by blahi 10y ago
I have experience scoring ~ 1TB daily. And a lot of smaller data sets spanning a few hundred gigs. It's not "hyper performant". Obviously doing things in scala or C++ will be faster. However rewriting the models would take mo
148.
by blahi 10y ago
Riiight. Everybody else is a bad engineer and you are the good one. With the single threaded R code... edit: The comment above has been extended quite a bit. Initially it was a single (abrasive) sentence. I still stand by my answer however.
149.
by blahi 10y ago
Do you have any knowledge about what Balmer's initiatives at Microsoft were? What his goals and objectives were?
150.
by blahi 10y ago
How much experience do you have in statistical computing, out of curiosity?

More ›