The primary improvement to human life expectancy was not medicine, but public sanitation. Prior to that, death was bi-modal, with peaks at infancy and old age. When people (and doctors) started washing their hands, the infant peak was flattened. Dead toddlers really kick the average in the nuts.
Quick, someone link that xkcd where a crucial open source technology is being maintained by basically a single person, while billion dollar companies rely on it.
Maybe, this is what people actually want to listen to and the algorithm is just doing what it's supposed to. I see people listening to these no name artists, the most boring unimaginative generic contemporary music. Maybe this guy just really knows what the average listener likes. In my opinion Spotify's algorithm is the best in knowing what I will like and probably will replay, I don't think it would work this well if they were manually directing us towards artists.
I feel like Spotify'a recommendations want to funnel me into a cohort whose music taste I supposedly share, but I actually don't. Last.fm seemed to both consider my likes more directly and more creatively. 15% or so of duds are fine, it's better than "oh you liked that dark techno track, it's mini-popular right now so have 20 more".
By the way, last.fm radio exists again, though YouTube is the music source these days.
I'm developing an open source app(flutter) I have already started it in a simulator(kvm). I just don't want to jump through all the hoops and pay to be able to publis the app somewhere for ios users.
People just want to be part of a community. Most of the time the only available open communities are church communities, so that's better than nothing.
Yes, I totally agree. I'm simply saying that you don't have to be uber religious to be a part of a church community. A good church should make you feel accepted no matter who you are.
There are perhaps exceptions, e.g. if you are intentionally trying to harm the church or members of it, but otherwise the church should accept you as you are -- "true" believer or not.
In practice you will probably find a lot of "bad" churches, so you'll probably have to shop around a little bit.
The RHEL sources aren't freely available. They are only provided to RHEL customers. Red Hat strongly discourages said customers from leaking the sources to Rocky, so Rocky won't say anything that might reveal where they got the sources. [1]
The GPL only requires that you provide sources to those to whom you distribute binaries, not to the general public.
[1] Red Hat can probably figure it out anyway. For example, they could slightly alter the sources and/or assets provided to each customer in a way that doesn't affect the functionality, and see which version turns up in Rocky's repository. But the leaky customer might too big for Red Hat to punish. Government maybe.
Just want to note that there is a lot of non GPL software in RHEL. e.g. Apache licensed or MIT. For which they do not have an obligation to provide the sources.
I already pointed that out multiple times to Rocky Linux staff. They have no answer to this. The whole concept of Rocky is a big bet, that Red Hat won't pull the sources of non-GPL binaries e.g. in cloud instances and other public accessable places, where Rocky is currently fetching their sources from.
If you choose not to participate in this gamble, then this distro may not be suitable for you.
> The GPL only requires that you provide sources to those to whom you distribute binaries, not to the general public.
Interesting. I looked at Wikipedia GPL page and saw this: "The GNU General Public License (GNU GPL or simply GPL) is a series of widely used free software licenses or copyleft that guarantee end users the four freedoms to run, study, share, and modify the software.[7] "
So any user of GPL covered software can share it with anyone. Right? Can any RHEL user share sources to Rocky? And to public?
I think this is an interesting question in general because GPL has also this clause
> You may not impose any further restrictions on the recipients' exercise of the rights granted herein
So I think it could be debatable if RHELs policy represents a restriction. Overall I feel this part of GPL has not been explored all that thoroughly and I feel it raises questions beyond this RHEL case. For example FSFs GPL FAQ states:
> For instance, you can accept a contract to develop changes and agree not to release your changes until the client says ok. This is permitted because in this case no GPL-covered code is being distributed under an NDA
I don't understand how that is not conflicting here; wouldn't your client be distributing the code to be modified to you, and that should be covered by GPL? And as such the NDA would represent additional restriction?
I suppose there could be a special case where the client would ask modifications for a publicly available sources and as such would not be distributing the code themselves, but I feel that can not be considered typical or general case.
Similar interesting case would be employees receiving copies of company internal forks of GPL code. Should employees have right to redistribute the code in accordance to GPL terms without threat of getting punished?
AFAIK mere exchange of code between employer and employee is not considered "distribution" for the purpose of the GPL. They are part of the same company, working on the same project. Similar relationships exist between client and contractor, client and lawyer, etc.
> The function of the contractor in such cases is nearly identical to that of an employee; however, because the contractor is not an employee, providing a copy of software to the contractor could be considered distribution. This is one of the thornier areas of GPLv2 interpretation, and it is discussed in more detail below
> A full discussion of the tenets of international copyright law bearing upon this issue is beyond the scope of this article, but it seems likely the question would have different answers outside the U.S [...] Therefore, the triggers for copyleft obligations, based on activity outside the U.S., may have a lower threshold than in the U.S.
They can, and they do, so the community eventually gets the sources. But Red Hat has every right to terminate their paid support contract with any customer who shares.
> Wouldn’t rebuilding from the RHEL sources be the easier approach?
Yes, if you can get the sources, of course. RH tells that they'll stop doing business with you if you share the source code you get from RHEL, which is Free and/or Open Source software.
> Also Alma doesn’t mirror Stream, they pull sources from several unencumbered places including stream.
Thanks for clarification. AFAIK they target versions of Stream and ABI of RHEL, but I learnt more today.
