It's hardly "a little more" though. We're spending almost twice as much and get almost half as little back. Objectively speaking the capitalist market has failed in this area. We have to first accept this fact before we can find any solutions. If we keep thinking "government is bad" and "private industry is good" we're only allowing further exploitation and pain; and for what? We need to be data/solution driven.
Ford twice as much I assume you're referencing healthcare specifically. Or healthcare system is horribly broken and in a terrible middle ground of government overreach and corruption while still technically being a private system. That's just about the worst example we could look at at which point twice as much doesn't sound nearly as bad for a worst case.
> Objectively speaking the capitalist market has failed in this area
Objectively speaking the non-capitalist systems have failed in this area too. The UK’s NHS is not capitalist and it is surely isn’t succeeding better than the US healthcare system.
You can find countries where “socialist” healthcare works, but you can also find capitalist healthcare that works.
> Objectively speaking the non-capitalist systems have failed in this area too. The UK’s NHS is not capitalist and it is surely isn’t succeeding better than the US healthcare system.
The NHS has been effectively privatized as much as the Conservatives could get away with. It is constantly underfunded. It is not an example of a system the government wants to make work. In fact making it not work is the goal.
Yet when a family member of mine needed a life saving operation, they got it promptly and it didn't send them into any debt, whereas in contrast there's something like 50k preventable deaths a year in the U.S. due to lack of healthcare and people of poor means actively avoid visiting the hospital, so the NHS still works better for the vast majority.
Yes, for us in the tech industry and for elective procedures it's not so great, but that also perhaps isn't what regular people need.
I love that term. If you prevent death one day, only for them to die the next, is that one preventable death? What if you do that five days in a row? Death is not preventable.
Extra years of quality life, I can get on board with. Especially because it sounds more manipulable, so you are more wary about it’s meaning.
To put 50k/annum in perspective, 3.5 million people died in the USA in 2021.
Don’t poor people get some free healthcare in the US - although relatively how good that care is compared with other countries I don’t know. I thought it was the middle class that lose everything.
the banking industry has been heavily subsidized by FDIC, the Fed, and bailouts. This corporate socialization is certainly worse than either fully capitalist or fully centralized.
The healthcare marketplace is hardly capitalist. It's deeply influenced by government policy and spending, as most healthcare expenditure is for the old, and it goes through Medicare.
If the US healthcare market isn't capitalist, then no country's healthcare system is.
It's rare to find a way to "no true Scotsman" as hard as communists, but you've managed it.
But to address your point: contrary to what you're implying, Medicare spending is actually lower on average relative to spending on 65+'s in other countries; in e.g. the NHS, they're 18% of the population and 40% of the spending[1]. i.e. per-person spending is much higher in retirees. Contrast this with the US where per-medicare-enrollee spending (13k) is almost the same as the average spending (12.5k), because the non-socialized (capitalist) part of the market pays so much more than in other developed countries, that it doesn't bring down the average despite the younger cohorts having many fewer health problems.
> If the US healthcare market isn't capitalist, then no country's healthcare system is.
Yes.
That said, developing countries are most likely to have one. If the government caps the number of doctors and hospitals in the country, like the US does with residencies and certificate of need laws, it's not very market-driven.
The government does not cap the number of doctors. There's no law that you can only have X doctors. What it caps is the amount of social help it's willing to subsidize. This is like complaining that the government is not "socialist" enough and therefore socialism is bad.
No, the US government caps the number of doctors because Medicare funds residency slots, and you have to get through one of those to get a license to be a doctor.
> There's no law that you can only have X doctors.
We have tons of these laws. All occupational licensing requirements are this law; that's the whole point of them.
The US government caps the number of residencies it is paying/subsidizing. Yes, you have to go through residency, but there's nothing stopping the private sector from taking on this cost (well, other than greed)
> Point me to one law that says you are not allowed to have more than X doctors.
Why can't we have doctors that other countries have agreed are qualified to be doctors?
Also, do you know how certificate of need laws work? It literally says you can't have a hospital unless the other nearby hospitals agree you can have it.
Taxi medallions are truly a "there can only be X many people doing this" system, but even if they're not written that way, these other laws are still intended to restrict the supply of healthcare.
nb my issue isn't whether the healthcare system is "socialist" or not, it's that it's bad. I don't think "just make the government pay for everything" is a solution to things costing too much though; if an ambulance ride costs $9000 the solution is to make it not cost that, not to share paying for it.
Your original argument was the government caps the number of doctors (not taxi drivers). I am not finding anything that would show this. There is a cap on the total amount the gov is subsidizing (should be increased, but that's a different topic).
If an individual is willing/able to pay for the cost of residency that the gov covers there's nothing to prevent one to do so. Of course, this is not a practical solution for most, but that's another story.
> nb my issue isn't whether the healthcare system is "socialist" or not, it's that it's bad.
Agree, no argument there
> I don't think "just make the government pay for everything" is a solution to things costing too much though
Of course not, but if the capitalistic / private market has failed, I'd rather try a more gov-focused approach (or something else) than let it fester and cause more and more pain.
> if an ambulance ride costs $9000 the solution is to make it not cost that, not to share paying for it.
Agree, but how do you do that when the "ambulance lobby" owns the politicians and media who then try to convince the population that this is "the best system in the world"? You can't find a better system when you already think you have the best one (not saying you do) or that the "other" system is worse because it starts with an "s".
These kinds of changes just make the OS look cheap (like a bazaar of sorts). They're visually ugly, but easily ignored (I rarely open the start menu).
What I'm more concerned / annoyed by is the trend of Windows fighting with you. For example, try to disable the real-time antivirus protection. The UI makes it look like it's off, but then it turns back on by itself. Then you go to group policy to make it explicit that you don't want it. It's off for a bit and then it freaking changes the group policy (resets it back).
First time I noticed it do this I though I got hacked because I couldn't imagine the OS doing something so malware-ish, but it does. What's the point of having a setting at all?
It's a neglected heap of legacy garbage, maintained just superficially enough to maintain profitability from its inertial marketshare (i.e., how everything runs on it already and how it's a massive coordinate problem to ever change that).
The KDE Dolphin file manager is way better than Explorer, feature-wise, and has been for probably over a decade. Dolphin was made for free by like five Dutch guys. Explorer is from a multi-billion dollar corporation that's buying every service and other company you know. I guess it's not profitable to improve user UX, so you have ridiculous artifacts like how much Explorer still sucks and how long it's sucked. :D (I think it only got a dark theme, like... 5 years ago? Linux file managers have had those for decades.)
If you're lucky enough to experience a change in your priority matrix such that things Windows is bad at (like programming) become more salient to your well-being than things Windows is good at (like playing AAA games), I recommend trying Linux (shout out to Manjaro) out. I'm glad I experienced that change, allowing me to flee to Linux and never look back.
Maybe time fro another, huge, anti-competitive fie from the EU, like when back when MS did all those shenanigans around internet Explorer? I wouldn't be surprised of we see similar proceedings in the next years around cloud services (everything from OneDrive to Azure and equivilants, Office, you name it) once the EU is done with mobile devices.
that's an interesting strategy to use alternate routes to the same information. Might be an opportunity to use more of the WIN+R shortcut to launch applications from that lil dialog. Makes me daydream of a day when Microsoft gets into a cat and mouse and starts slipping ads in many other dialogs that currently lack them.
Maybe not exactly elegant, but certainly much more neat and tidy, clean and consistent than it is now. From W95/NT4 up to at least W2K it was quite OK.
And using the old-style “theme” or whatever it was called, you could use that tidy and consistent UI throughout XP in stead of the garish “Fisher-Price” look, via Vista in stead of the annoying and distracting transparency, and well into Windows 7 in stead of whatever the “native look” of that version may have been like (I have no idea, switched to the old look first thing I did).
It only finally went to shit with Windows 8, and unfortunately restoring the Start Menu wasn't enough to make 10 good again.
I have to disagree slightly. The default for Windows 98 had a consistent aesthetic, consistent short-cuts across applications, and a sensible menu system (even if things were janky on the backend).
It's the only way we know of to stop greed/fomo running rampant. Folks need to be put in a time-out to cool off and force them to reprioritize. We like to think of the market as this calculated and rational entity. It's far from that, in fact it's closer to the opposite of that. The other option is to do price control, but that won't fly in capitalism.
There are other levers that can be pulled. Taxes are one example that could be much more targeted at the specific sources of the "greed/fomo running rampant". We instead chose to take it out on everyone.
I'd put taxation in the same bucket as price control because for this to be effective the taxes would need to be on the rich (in effect to drain liquidity) and that's as likely as price control.
I don't disagree with the unlikely nature of new taxes on the rich. That was basically my original point. It is wild that we decided that we would rather go into a recession than tax the rich and I don't understand why there is seemingly no public pushback against this.
There's historic precedence that increasing rates controls inflation. Yes, it's a very blunt and crude instrument. However, when you know this tool works, trying another tool is too risky to experiment with (in case it fails).
Inflation does have the risk of getting out of hand and being even harder to control. The risk analysis is that whatever pain increasing the rates would cause it pales in comparison to having hyperinflation.
As far as the lack of a pushback: My guess is a combination of apathy and the fact that it doesn't affect the average person quite as much as inflation.
I wouldn't say its the only way we know, rather its the only politically feasible move. Increasing taxes on the wealthy would arguably be a more effective way to take money out of the economy and wouldn't hurt non-wealthy people as bad, but good luck getting that through congress.
The only party that can act right now is the Fed, and the only move the Fed has in increasing interest rates.
The problem is that there's no proven good way to increase taxes on the wealthy. There's two flawed approaches: (1) increase income taxes, which doesn't tax the people who control the majority of the country's wealth, and (2) wealth taxes, which no country has ever figured out how to implement effectively.
> to do price control, but that won't fly in capitalism
Price control doesn't fly in any economic system. Putting a legal floor or ceiling on the price anything doesn't affect what it actually costs to produce it.
That depends on how authoritarian the regime is... However, I was talking about putting a price control over the level/amount of profit. If you look at the financial reports of late you'd see record profits during record inflation.
My understanding is that the .NET version is a port and only related in the style/approach. However, if I was them I'd consider renaming the lib. Looks like the name "Akka" is tainted at this point.
Since we have some PG devs here: Can we please have a way to reorder columns?
Coming from MySQL, this is one of the first missing features one hits. It's like moving to a new house where you're told you can never clean or move the furniture without moving to a new house.
It leaves the unfair impression that this is a "toy" db. "What other basic features is it missing if it doesn't have this?" Please don't think of it as trivial; first impression are very important.
Better to channel your displeasure into advocating for a future SQL standard to include this feature. Other DBs that support this do so via proprietary syntax extensions.
Technically it doesn’t require syntactic extensions, postgres stores the column index as the `attnum` attribute column of the `pg_catalog.pg_attributes` system table.
So this could be made to work by hooking into the storage system and rewriting the table and all pointers to the table when `attnum` is updated.
Without this rewriting this only works if the table was just created and has no data, and no external metadata referencing the columns themselves e.g. views, fks, indexes, defaults, rules, …
An alternative would be to add automatic packing (à la rustc) to postgres, decorrelating the “table position” and the “physical position” of the rows, this would also allow free “table position” reordering.
And while it’s by far the most complex option, one of the nice bits with it would be that the system columns could be packed as well. Currently there’s quite a bit of waste because there are 6 system columns (by default), the first 5 are 4 bytes, but the 6th (ctid) is 6 bytes, meaning 2 bytes of padding if your first column is a SERIAL, and 6 if the first column is a BIGSERIAL (or an other double-aligned column).
> And while it’s by far the most complex option, one of the nice bits with it would be that the system columns could be packed as well. Currently there’s quite a bit of waste because there are 6 system columns (by default), the first 5 are 4 bytes, but the 6th (ctid) is 6 bytes, meaning 2 bytes of padding if your first column is a SERIAL, and 6 if the first column is a BIGSERIAL (or an other double-aligned column).
FWIW, system columns aren't stored as normal columns in tuples. Some of them are implied (e.g. tableoid doesn't need to be stored in each tuple, ctid is inferred from position), others are not stored in the way normal columns are stored (e.g. xmin, xmax).
Couldn't an ordering column be added to the pg_attributes that determines in which way the columns are sorted when they are displayed? Standard SQL code can then be used to manipulate the display order of the columns.
There's like 5 different features wants I could add like this. A columnar database engine comes to mind (slightly more work involved) and poaching a few Sqllite QOL features.
Ultimately Postgres is a community driven product, and people want to work on what they want the work on.
I don't get angry that some kind soul hasn't volunteered their Saturday for free for me.
That's not exactly true. You can write your own patch for a feature and there's a good chance they'll decide they don't want it. It's their prerogative, but also slightly upsetting when you know they're blocking features you want. Just observe the bike shedding over a trailing commas patch [1].
There are lots of good reasons in that thread why the patch was declined.
The patch made things more lenient which makes things less compatible (this whole thread is about ANSI SQL standards and they do matter) and changes what was an error behavior to now silently succeed.
I read the thread and have to disagree with "lots of good reasons". I'm pretty sure there are plenty of things in the postgres dialect that are not cross compatible with any other dialect.
I agree with curating patches in general, but it's a double edged sword and inaccurate to say you can just spend your weekend to get something you care about.
Oh, absolutely; I'm just trying to raise the awareness/importance of this missing feature. Most devs that I know IRL that looked at PG had a very similar reaction to mine.
I'm receptive to the feature. But wouldn't it require locking the full table as it works? Seems like it would hurt availability of the biggest databases that would benefit most. Perhaps this is why it is thought somewhat impractical and therefore low-priority.
Depends on how the PG team decides to implement this feature. If they're OK with a virtual order (for "UI" purposes) while keeping the true order hidden then this would not require table locking (or it would be a very quick lock). If the order reflects the real / physical positioning then yes, it would probably require locking for a full table rebuild.
I like to have my `*_id` columns at the front, and `created_at` etc. at the back. And in the middle all the fields by descending importance. Just a personal habit so I can quickly lookup data in a table.
There is some small benefit to playing column tetris if you have columns with different sizes that waste space due to padding. I'm not convinced this would be worth the complexity of this feature, but in some cases reordering columns might have measurable benefits in reducing the size of the data.
I'm not talking about the physical layout of the data, just a thin UI layer that the DB tool could use. Maybe we could have two modes: physical vs UI ordering?
Views are not a good solution for this. The point is to be able see a specific order in 3rd party apps and when writing quick add-hoc queries (select *). Writing views for every table would just pollute the db.
Based on my understanding of DB storage (which is decades old and as I write the comment my explanation seems stupid but it’s what I was taught. Please someone correct me).
Image a table with the following:
id (int), title (varchar), created_date (date)
That data is stored in a similar order on disk (is this still true?).
So, if the initial insert has a title of 10 characters, then the date will get placed after those 10 characters on disk:
Int, 10 characters, date
Later, the row is updated with a longer title of 500 characters.
Because it’s longer, you now have to get more space for the title on another part of the disk.
Since there are columns after the title, it’ll either leave dead space where the old 10 character title was or it’ll need to move those other columns too.
I can’t remember which I was taught. I just remember that columns after variable columns get impacted when the data in a preceding variable column changes.
If the title were at the end of the row, the db could expand the data without needing to move other columns over (if there happened to be available space right there).
If you were updating with a shorter value, it could shorten the row by moving the terminating character to the new shorter location, freeing up the space the longer title was using.
Bottom line is you want to keep variable columns which get updated most frequently towards the end of your table structure.
Knowing which column that would be isn’t always clear when creating the table.
Again, this understanding was taught to me over 20 years ago. I’m probably remembering parts of it wrong and DB storage has likely (hopefully) advanced since then.
> That data is stored in a similar order on disk (is this still true?).
There should be no requirement for this. Columns in relations are not conceptually ordered, so it shouldn't matter for the things you're doing with the data anyway, and the database should be able to reorder the data in whatever way it likes, since desire to isolate the user from physical data structures was one of the main reasons for the rise of RDBMS.
1. That’s not exactly true, AFAIK all RDBMS return columns in table order when order is unspecified (`*`), and while they could reorder on retrieval
2. postgres definitely does not, and column tetris is absolutely a thing in the same way struct packing is (with the additional complexity of variable-size columns)
But the order of attributes in the presentation of the result relation has nothing to do with physical layout of base relvars -- primarily because any relationship between the two is purely coincidental. The vast majority of useful queries will not reuse the order of attributes in base relvars, so optimizing for the unusual trivial case by prohibiting better rearrangements that could be useful for a much larger number of use cases seems rather pointless.
And what PostgreSQL does is of course an implementation detail of PostgreSQL.
Why would you be trying to prevent the DB from reordering the data? You're not supposed to have better knowledge of what's good for your use case than an RDBMS that can collect usage statistics on queries and such. Ditto for compilers rearranging structures and such. When you start having hundreds of tables and thousands of queries, I don't see how you can do a better job than an automated system at that point.
DB storage is a lot more sophisticated than what you were taught.
Most databases use Slotted Pages to organize storage. Pages are fixed size and numbered by their offset within the database file. The page header contains the number of rows, followed by an array of offsets for individual rows within the page. Rows themselves generally are stored at the end of the page filling downwards. The storage engine can move around rows in arbitrary ways to consolidate free space.
Fundamentally there's no connection between SQL schema order and how table storage is organized on disk. For example in a column store there's often no contiguous row stored anywhere, instead there's just separate indexes per column.
Soothing my OCD is important, but it also helps when working in a team and having a shared ordering style. For example, we have certain types of columns that are always at the bottom. One would look there first. With PG you're forced to scan all the columns every time.
Ideally, postgres would play column Tetris behind the scenes and store the columns on disk in the most appropriate way, while allowing the representation to be changed at will.
With SQL Server the management tool does give you a way to do this. Yes, it does a table rebuild behind the scenes. The point is that it's easy. Don't have experience with the other two, but MySQL is the most popular so it kinda sets the tone whether we like it or not.
Well, writing a procedure that rebuilds the complete table in Postgres or Oracle is easy as well. I never needed this, but I am sure, there are some sample implementations out there.
Rebuilding the entire table doesn't seem feasible for large tables to begin with. Especially with a lot of incoming and outgoing foreign keys.
I disagree that MySQL is the most "popular".
It might be the "most used" one because of so many web hosting services included it for ages by default.
Tangential, but there are almost certainly more SQLite databases in existence than every other RDBMS put together, probably by 3 or 4 orders of magnitude.
I've worked with DBs for 20+ years as well. This is a quality of life type of improvement. If you've worked mainly with DBs that don't make this easy it's hard to know what you're missing. Do a search for column reordering for PG and you'll get a ton of hits.
The problem is that we hear a lot of different features touted as "the crucial missing one" -- I would definitely put this into the polish category, but items in that category are also important; especially for those with a MySQL background.
which recently has been picked up again -- that's awesome to hear
There's this constant tug of war between supply/demand (also applies to powerful/regular folks, those with money/without, etc). Inflation is the delta by which one side is winning against the other.
It doesn't matter what new model you come up with. As long as capitalism is around and pushing for ever increasing profits this kind of behavior will always be on the menu.
It's hardly "a little more" though. We're spending almost twice as much and get almost half as little back. Objectively speaking the capitalist market has failed in this area. We have to first accept this fact before we can find any solutions. If we keep thinking "government is bad" and "private industry is good" we're only allowing further exploitation and pain; and for what? We need to be data/solution driven.