Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Goodbye, data science

ryxcommar.com

Post reply on HN
Print view View original

291–300 of 415 posts

Re: Goodbye, data science

#291

Post by ghostbrainalpha »

daydream wrote: post #202
PragmaticPulp wrote: post #196

Earlier quoted context omitted.

Consequences often catch up slowly. It took years for Elizabeth Holmes to be sentenced because it takes time to collect evidence, build an airtight case, and give people their due process. As I get older, I'm actually noticing more and more consequences catching up with people, albeit slowly. The people I knew who drank heavily through their 20s and 30s are in much worse shape than basically anyone who made an effort…

> Consequences often catch up slowly. Agreed. Few people defy gravity; in the end most hit the ground. The phrase “slowly, then suddenly” comes to mind.

Should term consequences rarely align with expectations.

But Long term consequences always do.

Re: Goodbye, data science

#292

Post by fnands »

ketzo wrote: post #165
fnands wrote: post #131

Earlier quoted context omitted.

There is a bit of a joke that a data scientist is someone who can do better stats then the average SWE and can write better code than the average statistician. Both of those are relatively low bars to clear though

The way I heard the joke was "a data scientist is someone who's not good enough at math to be a statistician, and not good enough at programming to be a software engineer." Maybe a little harsh...

Harsh, but funnier than how I phrased it.

Re: Goodbye, data science

#293

Post by adammarples »

jstx1 wrote: post #25
adammarples wrote: post #20

Earlier quoted context omitted.

Depends. If their favourite data engineer says "Oh hey, I can write tensorflow too", then guess who get the job of to "productionizing" their crappy data science notebooks?

You have two much more likely options: 1. The person who developed the notebook is responsible for productionizing it. (No, it's not all crappy notebooks and some data scientists can indeed write high quality code). 2. You have someone like an ML engineer whose job it is to do this. What you're describing seems like the least likely option; at least on the teams I've worked on "I can write tensorflow" would get you n…

Well, anyways that's what I was doing this time last year. Much too small a team to have a dedicated ML engineer though.

Re: Goodbye, data science

#294

Post by jeffrallen »

> I have a sense that, if my current place of business needed to chop employees, that it would be a dumb decision to chop me over any data scientist.

Yeah, dude, good luck with that. That is emphatically not how layoffs happen. :(

Re: Goodbye, data science

#295

Post by tomrod »

glomgril wrote: post #276
tomrod wrote: post #154

Earlier quoted context omitted.

Agreed. I've run a "data science consultancy" in some form or fashion for three years now. When people say "data science" they mean one of three things: (1) MLE (2) Data Management (3) Data Analysis or Business Intelligence (applications of the same skillsets). (1) has a lot of ongoing innovation, be it in MLOps, autoML, mapping frontier ML to business cases, etc. Innovation is expensive if the investment strategy is…

Curious, what kinds of clients pay money for data science consulting? And does it feel like a sustainable business moving forward?

Like all consultancies, "it depends."

I've found three core customer sets that helped us define a sustainable business:

(1) government agencies (which tend to put most expenditure under labor categories, so they hire a lot of long-term consultants and contractors)

(2) mid-to-small sized non-technology firms that want better data science strategy or want to build data-driven features into applications/products (especially in novel ways)

(3) smaller technology companies that don't have the MLE and data management system capabilities.

My career has been in heavily regulated industries, so our customers often have an appreciation for the management and governance portion after experiencing negative data science outcomes from maverick types.

Re: Goodbye, data science

#296

Post by hnhg »

PragmaticPulp wrote: post #196
scottLobster wrote: post #156

Earlier quoted context omitted.

Not to get too off topic, but as a 35 year old engineer it seems the world in general has far fewer consequences than I was raised to expect. Everything from businesses with bullshit ideas flourishing at a loss, to January 6 even being possible (politics aside I expected the Capitol Police to crack a lot more skulls than they did once people started smashing windows), to the whole FTX situation and the tepid response…

Consequences often catch up slowly. It took years for Elizabeth Holmes to be sentenced because it takes time to collect evidence, build an airtight case, and give people their due process. As I get older, I'm actually noticing more and more consequences catching up with people, albeit slowly. The people I knew who drank heavily through their 20s and 30s are in much worse shape than basically anyone who made an effort…

I can't match up your anecdata with mine. I can think of numerous people who have done all the things you have mentioned and have no suffered no ill-effects. In fact, many have prospered from lying or cheating the system. From substance abuse to habitual lying, there were no consequences and actually in some cases great wealth was accrued. A great deal of awful people have a very fine life out of it, and there is no greater cosmic justice to address this.

Also, one could argue another interpretation of what you are advising is never take a risk, because it will have consequences. Well, in real life, it doesn't always. You can get away with a lot, and people do.

Re: Goodbye, data science

#297

Post by fnands »

SilverBirch wrote: post #284
fnands wrote: post #135

Earlier quoted context omitted.

Honestly, I can't tell you how many jobs ads I saw where I was wondering: "What would they expect me to bring to the table here?" Some companies just don't have the data, or heck even the need, for data scientist yet try and hire them anyway. Give smart people a fundamentally ill-posed problem and they won't get anywhere anyway.

It’s a great skill to walk in to a job and say “hey I’m the expert, that’s not a reasonable proposal, here’s the problem we can solve and here’s what we’ll do”. Much more value to the company, but hard to do.

Yeah I feel a lot of companies could do with running their problems past a consultant first.

Also, w.r.t hiring in cases like these, I think often the experienced candidates can smell that this won't be a good gig so don't apply, while the less experienced (or desperate) ones apply. This means the workers get stuck with an intractable problem, and the company gets stuck with workers who are too inexperienced to know better.

Re: Goodbye, data science

#298

Post by pinum »

jstx1 wrote: post #17
omega3 wrote: post #14

Earlier quoted context omitted.

> Data engineers don't work with machine learning at all. This very much depend on the company. From experience DE is used as a catch-all title.

Data scientist is much more catch-all from what I've seen. But a lot of that varies a lot by geography too (for example in the US people very often use DS very differently from how the title is used in the UK).

>in the US people very often use DS very differently from how the title is used in the UK

I haven't heard about this before and now I'm curious- can you elaborate on the differences?

Re: Goodbye, data science

#299

Post by tommykins »

I think is true for >general< data science houses and firms that offer Data Science across any domain - I'm generally wary of anyone who writes about Data Science as a concept, rather than the use of Data Science to solve a particular problem in a particular domain, the rise of the Machine Learning bros is very real.

Re: Goodbye, data science

#300

Post by SergeAx »

Lots of people are citing bad management as a most annoying thihg in tech. I wonder two things:

1) To recognize bad management, there needs to be awareness of what good management is. What is the common source of that awareness? How do one know good manager from bad one?

2) If good management is so crucial for tech company success, why there is no worldwide trend to help engineering managers become better? There's an awful lot of courses, bootcamps, learning videos, tutorials, git repos and so on for those who wants to be software engineers, but all I see for managers is self-help style books and articles, centered around typical situation "oh, shit, they appointed you to a managerial position, how do you cope with that?"

Post reply on HN