> PG argues that the crazy idea should be taken seriously predicated on the person having deep domain expertise
Depending on what you mean by "seriously". It's not that a crazy idea by a domain expert is certainly the future - but they might be right 1 out of 10 times, compared to a crazy idea by a nobody who might be right 1/1000. Given the potential large return of crazy ideas, it's worth the time to investigate in more detail.
Putting the version in a header looks "elegant" at first, as it keeps one URL for one resource. But in reality it adds accidental annoyance with no benefits over having the version somewhere in the URL.
* Have to set up response headers and caching carefully to make sure different versions are cached separately.
* A bit of extra complexity to set up load balancing.
* A bit of extra complexity to set up web framework routing to controllers.
* A bit of extra complexity for logging to track which endpoints were called.
* Calling the URL without specifying a version gets you some undefined version...
* ... or if you require a version header, can't preview GET endpoints in a standard browser.
* If different parts of the API have different latest versions, you can't encode it in an URL and therefore can't return URLs for linking between resources.
* On major version changes you might be removing, renaming or moving URLs, so why keep them pure and versionless in the first place.
Blows up a smaller percentage of the time when you forget it is often worse.
How often are your APIs changing incompatibly anyway?
When they do changing the URL should dwarf next to changing the surrounding code to the new data model. If that's not the case then what was the point of the upgrade?
My point is its pretty trivial either way. With headers, you can use the stripe style[1] time stamp version headers that give you a clearer picture of exactly targeting an API as it was at any given time, instead of as it was stated in the url (ie .../vN/...)
Kernel maintainers are volunteering their time and effort to make Linux better, not to be entertaining test subjects for the researchers.
Even if there is no ethical violation, they are justified to be annoyed at having their time wasted, and taking measures to discourage and prevent such malicious behaviour in the future.
Given the importance of the Linux kernel, there has to be a way to make contributions safer. Some people even compare it to the "water supply" and others bring in "national security".
> they are justified to be annoyed at having their time wasted, and taking measures to discourage and prevent such malicious behaviour in the future.
"Oh no, think of the effort we have to spend at defending a critical piece of software!"
I wish I had a dogecoin for every HN comment where a study is rejected because of a possible confounding variable that the researchers did control for and reported.
From the news article:
> The disorders were significantly more common in COVID-19 patients than in comparison groups of people who recovered from flu or other respiratory infections over the same time period
They controlled for the world situation, lockdown and even being sick with another illness. Of course the knowledge that you have covid could be stressful in itself, but that's also a result of covid.
> All cohorts included patients older than 10 years who had an index event on or after Jan 20, 2020
There were lockdowns on Jan 20, 2020? I thought lockdown measures basically eliminated influenza diagnoses—-were the majority of control cases diagnosed before lockdown measures were implemented?
There is no need to limit personal freedoms - anyone should be free to get the training if they wish to. But they don't have to remain employed as police afterwards.
> businesses like AWS that exploit FOSS to repackage it for profit
Exploit is questionable.
A true open-source project like PostgreSQL only benefits from Amazon and Google offering it as a paid, hosted solution. It makes it easier to adopt, sell to decision-makers, and gain more user and developer mindshare over alternatives.
If you are concerned with "competitors" offering your open-source project to more users, then perhaps being open source was just a poorly chosen marketing tactic that backfired.
> A true open-source project like PostgreSQL only benefits from Amazon and Google offering it as a paid, hosted solution
And like, how?
For a project to be exploited by tech corps usually the project itself needs to be already relevant and widely used across the industry. That's why it doesn't really need any publicity from AWS and that's why AWS "steals" it, their customers won't have to learn a new tool but can re-use their knowledge. Lower entry barrier = more customers.
> If you are concerned with "competitors" offering your open-source project to more users, then perhaps being open source was just a poorly chosen marketing tactic that backfired.
How free-markety of you.
Now imagine that Elastic took your suggestion. They never made the elastic suite FOSS in the first place but only sold it as a closed source product.
1) It would have never got widely used in the industry
2) We would have never got a free open source tool that helped kick-start many other companies.
You could say the same about Mongo. Imagine what impact it would have had on the startup world if Mongo wasn't a FOSS project.
That's why it is fundamental to protect FOSS projects from corporate exploitation, in order to protect the whole tech eco-system.
"1) It would have never got widely used in the industry"
Agreed.
"2) We would have never got a free open source tool that helped kick-start many other companies."
Disagreed. A different open source tool would have gotten the attention and that would have helped kickstart many other companies. Pretty much every open source tool or library has equally open source alternatives, sometimes even just as popular alternatives.
Elasticsearch is no exception, and its alternatives are about to get a surge in popularity.
"You could say the same about Mongo. Imagine what impact it would have had on the startup world if Mongo wasn't a FOSS project."
More startups would have used Postgres instead... Which probably they should have done anyway. Mongo only started to become usable for anything but raw streams of data, IMO, not long before it became source available. And even nowadays, I would still recommend Postgres (or any relational database) over Mongo for 95% of startups, even if it still were open source.
That said, I know that that is not the case with Elasticsearch, which probably is the best among its alternatives.
If the developer can get permission to build 3-5 story apartments on the lot, then their land suddenly becomes way more valuable than the single home ever was. Most retirees will gladly sell and live in luxury elsewhere. If a few don't, it won't make a difference.
If the developer can get permission to build 3-5 story apartments on the lot
wouldn't that only work for a single unit? Like, yes, if inside a single-family-zoned block a unit would get a special permission to be the only one to build high density, sure that piece of land would be more valuable than before.
But if the whole block gets rezoned, then there's a race to the bottom.
Many single family home owners will strenuously object to having a developer build a 5 storey apartment building next door because they don't want additional noise, traffic, and people looking down into their private back yard. It directly impacts their quality of life.
Depending on what you mean by "seriously". It's not that a crazy idea by a domain expert is certainly the future - but they might be right 1 out of 10 times, compared to a crazy idea by a nobody who might be right 1/1000. Given the potential large return of crazy ideas, it's worth the time to investigate in more detail.