It’s arguably more harmful to give a poor person $120K in debt to study something that almost certainly returns little, especially as those same people (statistically) may not know that some college degrees are worth significantly less than others.
This isn’t fully fleshed out, but the government could set a cap in loans based on anticipated the future earnings. Johnny gets into Med School and wants a $400K loan? Great! Iowa Central College wants to charge $200K for a dance major? Good luck finding students to enroll!
This also has the benefit of placing downward pressure on college tuition.
Businesses create value. That's not universal -- there's regulatory capture, monopolies, tragedy of the commons, etc. -- but it is fundamentally why capitalist economies have raised billions of people out of poverty.
Unions fundamentally redistribute the value that's been captured. That's not universal -- there's unions that make workers happier and result in more productivity, unions that are not adversarial and take on an HR function that's more in tune with employees -- but it is fundamentally why there can be successful companies with no unions, but of course no successful unions with no companies. Arguably, U.S. unions are more adversarial than European unions, and in many manufacturing, automobile, shipping industries have prevented their parent companies from adopting new technology, rolling out new practices, etc. to the extent that said parent companies were no longer competitive.
Workers labor mixed with capital creates value. Businesses are just a way of organizing that interaction that, if it succeeds, makes the interaction more efficient.
The businesses through the ownership of capital are the ones doing the redistributing of value, they are the ones writing the checks. Unions just negotiate that the redistribution allocates more of the pie to labor rather than the owners
I think we're roughly in agreement. Businesses are composed of the capital providers, the organizers (who decide how to deploy the capital), and the workers.
Capital providers colluding to set a loan rate is illegal. We shouldn't allow them to collude legally; they are a small number of experienced actors operating with information asymmetry already.
Organizers (i.e. senior management) colluding is--I don't know the legality--just unnecessary. They often are paid based on value add (equity, bonuses, etc.), so they are already reasonably aligned and need no further protection.
Workers can exist and do exist with or without collusion (i.e. unions). As a society, we've reasonably decided that colluding is legal, because workers often add a lot of value but have individual low value-over-replacement (hard to organize because of numbers, often inexperienced, etc.). Unions exist to help raise individual value-over-replacement closer to value-add.
What's being discussed here is the pluses and minuses of unionization. What I disagree with is people in this thread--not specifically you--comparing the collusion of workers to the existence of capital or organization (i.e. corporations are sometimes bad! why doesn't anyone discuss that!).
First, obviously people complain about corporations all the time. But second, private companies (capital and organization and workers) add value, and no society has done well without a large component of this.
There are many pluses to unions. But, especially as currently implemented, the minuses exist too, including that many of them focus on raising VOR but wholly ignore value-add, in a way that forces the business into slower and poorer decisions; or shielding some workers from the consequences of adding little to no value.
The capital tied up in any given business is colluding by definition because it moves in lock step. You don’t have multiple managers in a single org making conflicting business deals or bids with workers(for the most part, I’ve heard of a few companies organized this way). Hell even having multiple investors in a single company is collusion of capital owners when the company makes any sort of business deal.
Unions are just the workers colluding in the same fashion to increase their negotiating power
Whether you give me capital at a 1% interest rate of a 50% interest rate; whether you buy 1% of my company for $1,000 or 1% of my company for $100,000,000; are independently determined by each provider of capital. Collusion would be all banks working together to only provide me a loan for 30% interest even if it's obvious I can pay it back.
The use of capital is irrelevant to the question of collusion for providing capital. That's like saying all the workers on a factory line are inherently already colluding because they work towards producing the same car.
The amount invested in the company is provided by each separate investor, but those investors do not then negotiate separate deals with each worker. The investors “collude” together to creat the business and have greater negotiating power than they would on their own.
A union is merely the labor version of joining together to increase leverage and is not any special level of collusion greater than the investors is what I am getting at
And yes saying the workers are colluding because they are working on the same car is what’s happening if you are going to call banding together to negotiate rates is colluding. There is non union labor and other unions
Yes, perhaps all those scientists are idiots and you're the only smart person, even though you're too lazy to Google it before making a snarky comment.
Scientists just discover facts. Even small facts can be significant. How often have you seen "Obscure math theory that was widely accepted to be true finally proven"? The news media decides which facts are worth reporting on. I'm not disagreeing with this fact, but it's always good to remember that on a slow news day, the news media likes to overhype trivial facts.
Scientists usually discover probabilities, not facts. Mathematics is an essential tool for science, but most people don't consider it a science itself.
Hi! I used Wanderlog to plan a recent month-long group trip, which was definitely the most complex vacation I've had to plan. For context I am very active when traveling (e.g. multiple activities each day); so not sure how my experiences map to others.
The best part of it was (going to a foreign country) being able to find / identify all the attractions relative to each other, so I could go to cluster A on Monday, cluster B, on Tuesday, etc.
The hardest part of it (and why I needed to create a separate google sheets anyways) was--once I figured out opening hours of different locations, hard-to-book activities with limited reservations--the ease of moving things around more fluidly e.g. cluster B on Monday, cluster A on Tuesday, etc. and having a more information-dense view so I could see larger portions of the itinerary at once.
It would be cool to have an "input everything" --> "input time restrictions / unmovable things" --> output planned activity cluster type workflow.
For this to be usable to me (level of knowledge: I can do all of what's done on this page in Python / R, but don't have a PhD in stats or anything), I would need:
- Some sense of how training and validation is done
- Model weights
- Something that helps interpret fitting / overfitting
I'm not sure it's super useful for someone below my level of knowledge (or maybe at my level but just don't know Python?). It seems like a random marketing person, shopify person, etc. would need:
- Better understanding of when to use regression vs. classification
- Help interpreting of whether the MAE / loss is good or bad
- Some automatic way to prevent overfitting
- Guidance on what consitutes good data and how to structure it for input
- Examples of how it might be applied to their use case
- Knowledge of how often models fail / how much they should be indexing on the model
1) The target user is: a freelance marketer, small/medium enterprise without dedicated data scientists, technical people (e.g. engineers) from other fields, or without extensive stats/ml knowledge.
2) RE:lack of hand-holding.
This is likely the biggest challenge for this project: showing people how to think about ML, and how to use it to derive value for their business, without going through hours of training or lengthy tutorials.
> - Guidance on what consitutes good data and how to structure it for input
> - Examples of how it might be applied to their use case
Most users are stuck at this initial phase (preparing the data).
One thing I'm working now is adding use-case focused guides: short article explaining how a realtor would go about building a model to help them roughly value houses (including data collection). I hope this helps with these two points.
> Help interpreting of whether the MAE / loss is good or bad
There's a couple things I'm working on that might help:
1) Show metric improvement relative to a baseline (e.g. MAE for a model that always predicts the mean).
2) Show both train and test curves. The current curve is only on test data.
> Better understanding of when to use regression vs. classification
> Some sense of how training and validation is done
I'm currently redesigning the UX around a step-by-step flow (for initial users at least), that should give a bit of room to explain things along the way (e.g. what classification/regression means for total beginners).
> Some automatic way to prevent overfitting
Medium-term models there'll be a mode to continuously train models to tune hyperparameters, that should help avoid overfitting. Until then it's mostly handpicked parameters (including regularization), and having tested this on Kaggle challenges it still sometimes beats my hand-written ML code :)
> Model weights
You can already download the model weights (download icon next to the model name) ; or do you mean feature importances? That's a planned feature, but it's not straightforward to implement in a generic way so might take a month or two before it's shipped.
Ah, for what it’s worth I looked for a while and didn’t see the download icon until you told me it was there, and I consider myself pretty good at picking up new user interfaces relative to your average user (e.g. became competent at Photoshop, Excel, etc. without handholding). It wasn’t intuitive that’s where the model weights would be stored so my brain never looked for a download icon.
A bigger issue is I realized the weights downloaded do not include the data preprocessing... So proper model export will unfortunately take more work before it's fully ready to use.
As a random "marketing" person, I agree. I can blindly fumble my way to train a model, but a little more hand-holding would help me understand context and instill more confidence in the results I magically get by clicking things.
Agreed, although the app is still much easier to use than other ML apps I've tried, it's still a bit too confusing/"magical" for non-technical folks.
As mentioned on the parent comment, this is the #1 priority, and I hope a redesigned flow (e.g. step by step from loading the data, picking the target, then features, and explaining things at each step) will help here. I'll also have a page with concrete use cases, including marketing.
Please reach out (email on the website) if you have some ML use case you'd like to solve with ML Console, happy to help you prepare your data etc.
I am not aware of any data that colleges--liberal arts or otherwise--teach much in terms of critical thinking. Rather their value seems to derive primarily from their ability to select talented students and provide them with a network of similarly talented peers. That and they get "credit" for the learnings of students as they age from 18 - 22/26, when that's a pretty ripe time for maturing thought with or without the classroom.
Moreover, it's clear that most colleges agree with my viewpoint. For example, if Harvard's product was an amazing curriculum, they could expand that to many, many more students than their current class size and charge for it (instead, their actions are rational when their product is exclusivity and high-talent networks).
Though I have a degree in math, my undergraduate experiance was at a liberal arts college. Outside of my major I was required to take at least 2 history classes, 2 philosphy classes, 2 theology classes, 2 lietrature classes, 2 social science classes and a foreign language (modern or classical) to the intermediate level. I was a very rich and rewarding experience, and I firmy believe every bit of it makes me a better software developer and human being.
That being said, there was no explicit 'critical thinking' aspect to it, rather one learned the importance of reading and writing. If we define 'critical thinking' as the ability to bith understand a subtle argument and to make one, it was the reading and writing that did that.
Similarly my BA required spending the entire senior year writing a thesis paper requiring much slowing down, critical thinking and rethinking. I agree with OP’s assertion but it probably doesn’t apply to every college.
CEOs are generally not going to take on any risk to the company -- especially one that exists in a highly regulated segment of the market -- for the purpose of an interesting interview. I would be surprised if the author genuinely expected to get an honest answer from the Pfizer CEO about whether the FDA is doing a good job.
Unfortunately, there's many corporate environments where technical staff are managed by non-technical staff. This comment isn't debating the merits of that, but when the manager (or especially their manager) doesn't really intuitively understand the difficulty of your work or how much you've accomplished, they measure by other things like "responsiveness".
This is similar to how patients will rate doctors (lawyers, auto-mechanics, electricians) higher by things like waiting room design, ability to make small-talk, willingness to give antibiotics, etc. which are important but actually do not correlate with better care.
So I've found that optimizing for "responsiveness" is important in and of itself, and much easier than the alternative of "let me make sure non-technical senior management person X spends the time to understand my contributions at a deeper level".
Yep. Stuff like this is the reason that I moved from the technical side into management about 4 years ago. Seeing the dangers to my team from managers who don’t understand the challenges and constraints became a situation that I couldn’t take anymore.
Now, since going that route I’ve gotten deep into SAFe (Scaled Agile) because I believe that it truly solves so many problems. SAFe recommends using “5 whys” in certain situation and SAFe itself isn’t without valid criticisms either. IMO it’s still by far the best option for a long list of reasons.
Like anything in the wrong hands though, it can be co-opted. Everything revolves around good leadership.
Agree 100%. Many riots / burning of buildings / lootings were allowed while media spurred them on and called them peaceful protests. Police in many cities gave up on enforcement.
In those scenarios the state no longer has a monopoly on violence.
