Hacker Newsnew | past | comments | ask | show | jobs | submit | jholdn's commentslogin

My experience with consulting firms is there are two paths for high performers. Either you move into a sales/relationship focused role (which there are fairly limited slots for) or you move client side and probably buy from your old employer. The good workers do well either way (when I say move client side, I mean to high level positions - importantly one with control over a budget to afford consultants) and the consultancy makes money both ways.


This: If you want to climb the ladder, the no1 way to becoming a partner is .. SALES .. SALES .. SALES, regardless what you selll: As long as you bring in money to justify your business unit/cluster, everybody is fine. Just sell more hours - thats the only goal.


This is somewhat to credit AI model design. I think this is how you'd want such models to behave with esoteric subject matter. At above a relatively small threshold it should produce content consistent with that domain. Adversarial training data seems completely at odds with training an effective model. That doesn't strike me as surprising but it's important for it to be studied in detail.


This program sounds great and I think we should incentivize it. Unfortunately, I think it requires a constitutional amendment to work. We can’t rely on well meaning administrators to overlook the slave labor exemption for criminal punishment; these things will be exploited.

I guess with knowledge work there is some protection because it’s hard to force. Though, it would be desirable to extend such programs into other forms of work.


One hypothesis I don't see mentioned is that time limited roles pay more. For example, a person brought into to oversee an acquisition or transition is going to be paid phenomenally well (and, I think, rightly so). But part of that job is making your own position redundant. If you're still there after a few years, you aren't doing your job. I expect there are many other examples, roles that are project based and become redundant when the project is completed. And, I expect, many of these pay more than operational jobs, taking care of some day to day tasks of the business. Probably rightly so; if the projects successful, I would think building something new would create more value as it should produce something that continues producing value into the future - higher risk, higher reward.


Depends on the business. If all revenue is coming from contract work then everyone's job is tied to having a contract. If revenue is streaming in from services then most jobs are tied to support and are buffered by much higher margins.


I believe healthcare as a business (in the US) relies on the fact that the utility function of healthcare services become infinite. All other utility relies on an individual's good health. As such, it becomes completely determined by wealth distribution, not the value or cost of the product being traded. It's natural that a healthcare system that attempts to follow market based solutions follows the money. In the US, all the money has moved to cities so healthcare is following.


I think it’s just a phrasing issue. Should be: One of them incorrectly used the width of the new line character in its calculations, and so the box that it made was too small [for the text size computed by the other].

I agree it makes it sounds like the box sizing was wrong when it must be the other way around.


Isn’t there also a component that PE is incentivized to work for existing investors more than new investors. If they mark down an asset in a bear market and a new investor puts money in they will potentially dilute future profits for existing investors by possibly letting them in at a lower price. If the market recovers, the new investor sees all those gains. If the asset isn’t marked down, those gains are shared with existing investors (b/c the new investor had to pay above market).


The majority of private equity (incl. VC funds) funds are close-ended by definition [1].

Once the fundraising completes for a fund, no “new” money is allowed into the fund. New investors can invest with a PE firms in 2 ways: invest in a new fund, or purchase shares of an existing fund on the small secondaries market [2].

So no, in most cases PE portfolio managers are not going to track “new” vs “existing” money within a fund.

There are exceptions, ie there are PE asset classes that are open-ended (ie. The infamous Blackstone REIT [3]). Open ended funds are more common for public equity hedge funds.

[1] https://www.investor.gov/introduction-investing/investing-ba...

[2] https://www.institutionalinvestor.com/article/b1zspcywbpn2h7...

[3] https://on.ft.com/3Zc7Wku


I don't think this is correct usage of the term "unfunded mandate". "Unfunded mandate" means when the federal government requires a separate entity (e.g. state, municipality, company) to perform a function without reimbursement or appropriation of funds. Executive orders (that don't rely on special powers granted to the executive by legislation like emergency powers) only apply to the federal agencies. "Unfunded mandate" cannot describe mandates the federal government places on itself.

edit: You're right, what's required is a funded mandate. Gas is too foundational to the function of the country to leave to a private enterprise that isn't (and shouldn't be) responsible for a guarantee to provide their service. There should be federal intervention by buying a stake in the industry (though maybe in the case of the petroleum industry the federal government already pays enough but I say that from meme knowledge not actual knowledge).


I am open to more appropriate terminology to indicate an executive order is set forth with no funding provided alongside.

For these orders, the President’s Budget should include requests for Congress to provide appropriations for the work that will need to be performed (the delta between current state and expected state).


Within various parts of the US executive branch, presidential executive orders that say "do X" without funding to match are also called "unfunded mandates" - at least sometimes. I certainly heard the phrase used that way. I don't know if that's an abuse of the term or not.

It is true that it's different when it's applied within the executive branch. The US executive branch has to obey the laws, but it does have some discretion on how it prioritizes the work. So, for example, it could prioritize some of these activities by de-prioritizing (and thus reducing the funds) of other activities. That's hard, because it's often harder to stop something you're already doing (there are often existing groups in place who will be unhappy if you stop it). But it's officially possible, and if it's important enough, then it can be done. The US government's executive branch bureaucracy cares very much about what the president orders, though when there's limited funding the work must be prioritized.

Some of these steps, such as creating new FAR clauses & adding them to contracts, don't require new money (though their results may impose new costs later). But it's fair to say that others will cause costs, and they'll need to go to Congress to try to get that funding.

But that's not insane. The attacks on SolarWinds and friends have been estimated to cost $100 billion in damage to the USG (that's probably too high, but it shows real harm). The latest Colonial Pipeline ransomware, which threatens the US east coast, is making the news. Congress lives on the US east coast. I can easily see that money flowing from Congress.


Agreed, and I see your point, I just see it as the EO doesn't have the authority to (and thus doesn't actually) require such things from private companies and not that it requires something without funding.

In terms of what should be done, I'm not sure whether companies should be expected recoup their costs by raising prices, reducing margins or expect federal funding. My gut reaction is that this level of security should be a cost of doing business and not require federal funds.


It's an interesting data discrepancy - fewer smokers in covid-19 infected individuals - but it doesn't tell us much. The survival bias is a problem. They took the smoking status of alive patients, not in the ICU. The results could indicate the opposite effect. Smokers tend to be in the ICU or die at an accelerated rate so are under-represented. That said, one would expect counter-acting effects in such a situation that would make the counter-intuitive result of fewer smokers meaning it's worse for smokers less likely, e.g. greater diagnosis rate in smokers due to increased severity, higher rates of infection in smokers, etc.

I don't think any conclusions can be drawn from this study. It does suggest that there is value in investigating the impacts of smoking on Covid-19.


How about I use a simple suffix tree instead?


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: