These are chat archives for FreeCodeCamp/DataScience

29th
Sep 2016
Darwin RC
@darwinrc
Sep 29 2016 01:50
@jomcode @evaristoc take a look a this researcher's blog: http://evelinag.com/
John L.
@jomcode
Sep 29 2016 01:51
thats a really useful link thanks @darwinrc
CamperBot
@camperbot
Sep 29 2016 01:51
jomcode sends brownie points to @darwinrc :sparkles: :thumbsup: :sparkles:
:cookie: 425 | @darwinrc |http://www.freecodecamp.com/darwinrc
Darwin RC
@darwinrc
Sep 29 2016 01:52
@jomcode @evaristoc F# is a kind of ML (the functional language) derivate, while Scala is a mix between OOP and FP
Selection_001.png
Darwin RC
@darwinrc
Sep 29 2016 01:59
@jomcode from my little knowledge about FP, I would choose Scala for "real world" scalable applications and Haskell to learn the "pure" functional approach. I don't know much about F# but I learned SML long time ago and used it for some VR applications
John L.
@jomcode
Sep 29 2016 02:00
what does SML stand for, Mathlab?
nm just looked it up
Darwin RC
@darwinrc
Sep 29 2016 02:04
no, it is a functional language by itself
standard ml
John L.
@jomcode
Sep 29 2016 02:04
gotcha
Darwin RC
@darwinrc
Sep 29 2016 02:09
@jomcode finally if you wanna look deep into the roots, Lisp is the one
John L.
@jomcode
Sep 29 2016 02:10
haha im not trying to be a purist or anything. most of the material i look at for learning is based on haskell but i implement it in other languages. f# seems like a nice middleground for being able to be productive
Darwin RC
@darwinrc
Sep 29 2016 02:16
Here is another researcher from MS https://lenadroid.github.io/posts.html
Hèlen Grives
@mesmoiron
Sep 29 2016 07:44
@erictleung thank you; I'll focus especially on the right questions as that skill requires exceptional thinking and creativity. My interest in machine learning stems from my deep interest in languages as communication means.
CamperBot
@camperbot
Sep 29 2016 07:44
mesmoiron sends brownie points to @erictleung :sparkles: :thumbsup: :sparkles:
:cookie: 415 | @erictleung |http://www.freecodecamp.com/erictleung
Hèlen Grives
@mesmoiron
Sep 29 2016 07:48
@evaristoc that's why I joined the bandwagon. Little sister watching Big Brother. It is important to have people who bridge the connections. People are still superior when it comes to trends spotting as this requires the skill of understanding the minute in big tails as to speak.
evaristoc
@evaristoc
Sep 29 2016 08:50

People

By the way... (@erictleung)

Did you know that MS is now the owner of R-evolution Analytics and R Server?

MS, as many other big ones, is investing a lot on acquiring Open Source technology nowadays. The last year was the time for R-evolution Analytics. I was actually unaware until I found that the logo of a training I took with them 3 years ago changed into MS on my linkedin just few days ago.

But this is important. For those who don't know what R-evolutions does, they were mostly focus on on the Big Data applications and libraries of R and a sort of Nonprofit consultancy/certification organisation, just like Cloudera is for Hadoop technologies.

Much of the current Open Source technologies will stay so, but in the future certification will be the way to go, just like it has been for MySQL (first Sun Microsystems, then Oracle). It happens that for many big organisations acquisition is a way to kill the innovation, not necessarily by avoiding its use but by deterring continuity and slowing down momentum in favour of their in-house products.

By the way, who have heard about MySQL since the takeover by Oracle? Is MySQL still in use? Nowadays Nonsql is more popular though.

Eric Leung
@erictleung
Sep 29 2016 08:58

@evaristoc I didn't know that it happened formally! That explains why this website exists https://mran.revolutionanalytics.com/. I got redirected to that site while looking up packages in R and was a bit confused. Makes sense now. Thanks! :+1:

I agree non-SQL technologies are more "popular" but probably more production systems still use standard SQL technologies (based on opinion, I don't have any evidence for that claim :smile:)

CamperBot
@camperbot
Sep 29 2016 08:58
:cookie: 314 | @evaristoc |http://www.freecodecamp.com/evaristoc
erictleung sends brownie points to @evaristoc :sparkles: :thumbsup: :sparkles:
evaristoc
@evaristoc
Sep 29 2016 09:04

@erictleung yeah... just confirming your comment:

probably more production systems still use standard SQL technologies

Also checking some interesting reports on Internet and found that actually what it is happening is that Open Source products might be eating the market of proprietary ones in the case of relational databases.
http://www.infoworld.com/article/2916057/open-source-software/open-source-threatens-to-eat-the-database-market.html

At least a projection in 2015. So I think the big ones are just for the time being trying to capitalise through certification and post-services, pretty much exploiting the open source capitalization model, I guess. I haven't paid attention to this one though...

evaristoc
@evaristoc
Sep 29 2016 09:23

Other thing could be trying to reach different markets. Open Source technology is king when it is about emerging start-ups, but those start-ups usually shift into proprietary technology once they have reached a certain level of maturity. Why? They need to pay less attention to techs and more to sales and clients. For some of them focusing on tech could be a distraction at some point. Also, big providers of services (like IBM) are usually big clients (like for example a Marketing Consultancy company). A company might adopt a proprietary product not only because will be well serviced but also because "it is the product of my main client or investor (let's say MS)".

As an analyst I find very interesting how all those companies are trying to make business... the hard part of working on this is the political part, that part the data can not tell you about, the secret meetings between top CEO's, that phone call, the confidential email or the results of an conversation in a dinner or bar. When it comes to analysing this kind of things, most of the time analysts presume to have a lot of data to bring conclusions, but there are unexpected actions or decisions we are not always aware of, really.

Alice Jiang
@becausealice2
Sep 29 2016 12:57
Just finished watching this presentation on 'hacking' a career in data science and thought I'd share here for anyone who might need it or have an opinion they'd like to share on it. :)
Hèlen Grives
@mesmoiron
Sep 29 2016 13:04
@alicejiang1 cool
Darwin RC
@darwinrc
Sep 29 2016 13:15
thanks, @alicejiang1 ... watching
CamperBot
@camperbot
Sep 29 2016 13:15
:warning: could not find receiver for alicejiang1
darwinrc sends brownie points to @alicejiang1 :sparkles: :thumbsup: :sparkles:
Darwin RC
@darwinrc
Sep 29 2016 13:18
It seems there are many data scientists from that region of the world: Gabasova, Dzenisenka, Ivanova :smile: :+1:
evaristoc
@evaristoc
Sep 29 2016 14:48
@alicejiang1 :+1:
@darwinrc :) :) :) You might be joking but it is actually happening: Eastern European distinguish for their strong emphasis on pure maths and strong theoretical background, for what I have seen. They are really good at it.
Darwin RC
@darwinrc
Sep 29 2016 14:54
@evaristoc you are right. In that matter they look ahead of India and China.
evaristoc
@evaristoc
Sep 29 2016 14:55
I think so...
Don't tell that to our Indian and Chinese friends here!
they are also good...
Darwin RC
@darwinrc
Sep 29 2016 14:59
Hahaha, of course they are :+1:
evaristoc
@evaristoc
Sep 29 2016 15:01
:)
evaristoc
@evaristoc
Sep 29 2016 20:25

@alicejiang1 interesting the Ivanova... Gabasova... Dzenisenkova... Tetiana's talk.

Veeery clever I must say... and sooo convincing... although the last question comes from someone that although bitter has a point.

Actually: I would be VEERY careful in taking for granted her arguments to explain the trends. Not sure about how she handled the PhD discussion, for example... and University: well: it could be the most important signal belonging and networking point of your WHOLE life (eg. MBA's are based on that argument, basically).

Unfortunately she was very convincing but I believe she got a bit away when making her case. But good for her for being a REALLY clever person.

Alice Jiang
@becausealice2
Sep 29 2016 20:26
@rleva
Idk what that was
@evaristoc she did say at the top that this was anecdotal and based on her experience. So her presentation was based on a single data point which was biased for plenty of reasons. Not to mention she claims it only took her six months while also explaining how there was almost a year prior to quitting her job.
I just wanted to give people without degrees some hope :)
evaristoc
@evaristoc
Sep 29 2016 21:22
Good, Well Done!! (I don't have a degree)
Actually we should some of us make a hashtag in twitter like that: #Idonthaveadegree
Good night people!
Alice Jiang
@becausealice2
Sep 29 2016 23:26
When I get a job I'll start that hashtag
Jacques
@razerh0
Sep 29 2016 23:44
Hi Everyone. Glad to be a part of this wonderful and so far interesting chatroom. I found my way here through Code Camp Medium.
Darwin RC
@darwinrc
Sep 29 2016 23:59
Welcome @razerh0