portsnap being retired - what's the alternative?

Status
Not open for further replies.
From that post mentioned by 20-100-2fe:
"Desktop is not our focus. Servers are." But no one is going to run a server OS they don't know about.
People run Linux because they know it. They use it at their home and since it is good enough for servers, they started to use it there as well.
You know this is my mantra. Spread the word!
 
If they consider Git as a collaboration to, it means that for them Git = github.
Github has collaborative features - e.g. ticketing, wiki.

No, that's an entirely nonsensical take on it. FreeBSD already has a bug tracker, a wiki, and a patch review system in place. They are not going anywhere. The better collaboration part refers to Git being specifically designed around workflow where vast majority of changes are submitted by people without direct commit access. Git is just that much better at dealing with patches than SVN.
 
msplsh You seem to be confusing wants and desires. No one who doesn't care about FreeBSD is going to fix bugs on FreeBSD just because there is a bug and just because it uses git. Yes, it's possible to write for FreeBSD and never use svn. That was never a discussion point so I don't know why you brought that up.

The only point I see in using git is (if) it solves problems that are serious enough to make one switch. One of the talking points I've seen is that it makes FreeBSD visible on github but I don't think you need to use git in order to make yourself visible on github. In fact, when one wants to get a copy of software there, github itself offers the choice of downloading a zip file instead of cloning with git.
 
Git has real qualities that should not be overlooked. It makes it very easy to work with subcontractors, for instance. Or to create derived works from open source software while keeping the possibility make either side benefit from the other's enhancements. Or to safely and reliably manage what will be put into production until the very last moment.

From the developer's perspective, it is easy to use and self-documented. This means it is pretty usable even without a GUI.

These are all very good reasons to adopt Git, but there is no polite way of qualifying the migration to Git in the hope it will attract new developers.
It would certainly be more profitable for FreeBSD to wonder why contributors leave or step back, but for the above reason, this will not happen.
 
[...] It would certainly be more profitable for FreeBSD to wonder why contributors leave or step back, but for the above reason, this will not happen.
You mean the lack of a vision for the OS? EDIT This would be off-topic & worth it's own thread IMHO.
 
You mean the lack of a vision for the OS? EDIT This would be off-topic & worth it's own thread IMHO.

I have a few non-technical talents and discourse analysis is one of them.
I've thus immediately noticed the article referred to by msplsh is quite similar in form and content to corporate and political discourses.
This is clearly not what I expect from the core team of a large open source project.

So to answer your question, it's not just a matter of vision, it's much deeper than that and, as it cannot be changed, not worth discussing if in the hope of improving things.
However, FreeBSD is true open source software ; if someone feels like creating another branch in the BSD family tree and deems the effort worth being made, it's possible.
 
  • Like
Reactions: a6h
[...] I've thus immediately noticed the article referred to by msplsh is quite similar in form and content to corporate and political discourses. This is clearly not what I expect from the core team of a large open source project.
So what do you expect from a large Open Source project?
So to answer your question, it's not just a matter of vision, it's much deeper than that [...]
Which is specifically?
 
Ports index is required for "make search". Also I think is used to speedup "pkg version -vL=" but you can also use this command without ports index. Ιs it used elsewhere?
 
Ports index is required for "make search". Also I think is used to speedup "pkg version -vL=" but you can also use this command without ports index. Ιs it used elsewhere?
egrep -ril '(index|ports)' {,/usr/local}/etc/periodic/*
Code:
/usr/local/etc/periodic/monthly/300.statistics
/usr/local/etc/periodic/weekly/400.status-pkg
and e.g. ports-mgmt/portrac and tools alike might use it.
It's also in pkg.conf(5), so some other pkg(8) tasks/commands might access it.
 
Git has real qualities that should not be overlooked. It makes it very easy to work with subcontractors, for instance. Or to create derived works from open source software while keeping the possibility make either side benefit from the other's enhancements.
That's for when those subcontractors are using git. That doesn't mean FreeBSD needs to use it. Subversion has great qualities, too, like centralized version control versus distributed version control which git is. This is another thing that bothers me.
 
No one who doesn't care about FreeBSD is going to fix bugs on FreeBSD just because there is a bug and just because it uses git.

I don't know how to make a bugfix to FreeBSD but I do know how to make a PR on a git repo.

Yes, it's possible to write for FreeBSD and never use svn. That was never a discussion point so I don't know why you brought that up.

The point of bringing that up is that svn is not integral to FreeBSD.

What's the point of beating this dead horse? They're changing over and since most of the people on here don't commit to the repo, then I'm not sure why our opinion matters. They're changing the bus from gas to electric. It runs the same route, why all the squabbling...
 
So what do you expect from a large Open Source project?

In short: to be driven by a mindset as open as its sources.
This implies that it finds the justification of its existence outside of itself.

Which is specifically?

The same as the other members of the BSD family (and as Illumos): each of them is a cocoon and touching any single thread of it, even to make it more comfy or look nicer, is considered a vital risk by its inhabitants. They just want it to stay the same until they die, full stop.

This explains the Git story. On one hand, the CT says they decided to migrate to Git to attract new developers, but on the other hand, as it is complete nonsense, everybody in the FreeBSD community understands this will not happen, so everybody feels safe, nothing is at risk of the slightest change.

The only problem with this is for outsiders: they come in, find great things here and less great ones, they'd like to contribute and they discover they've been deceived by an insincere discourse.

Being familiar with systemics, I understand the situation and feel no resentment, but I've already noticed that some people feel quite sad and bitter about it.
 
  • Like
Reactions: a6h
It's not like "you buying it" changes anything. I'm just providing the rationales.
Certainly, I've stated in my previous posts on this and the other threats, that my opinions on these subjects are irrelevant. but I retain the right to keep whining.
 
We're drifting off-topic :) But I like this subject. Maybe we should move over to another section (off-topic) or IRC (which I still not have configured...)
In short: to be driven by a mindset as open as its sources.
This implies that it finds the justification of its existence outside of itself.
I can not comment on the mindset of the CT people. I do not know them. Do you? FreeBSD is widely deployed to drive services for universities, governmental & NGO's agencies/offices, companies etc., companies like Netflix, WhatsApp, NetApp etc. use it as-is or building blocks/foundation to build their own OS, and it has a healthy (but small) user base. Derived projects like user-friendly firewalls & NAS have a good reputation. I'd call that a justification of existence. It's written out as "to provide a BSD OS with focus on robustness, stability & performance". There seems to be a demand for such, and FreeBSD delivers in a fairly good quality IMHO.
The same as the other members of the BSD family (and as Illumos): each of them is a cocoon and touching any single thread of it, even to make it more comfy or look nicer, is considered a vital risk by its inhabitants. They just want it to stay the same until they die, full stop.
The users mentioned above keep demanding new features, and they have vital interest to get these implemented. I can not see any stagnatation. Maybe some progress does not happen fast enough, but there is progress.
This explains the Git story. On one hand, the CT says they decided to migrate to Git to attract new developers, but on the other hand, as it is complete nonsense, everybody in the FreeBSD community understands this will not happen, so everybody feels safe, nothing is at risk of the slightest change.
As I understand it, the main driving force has been limitations regarding the developer's workflow, which git seem to solve better than Subversion.
The only problem with this is for outsiders: they come in, find great things here and less great ones, they'd like to contribute and they discover they've been deceived by an insincere discourse.

Being familiar with systemics, I understand the situation and feel no resentment, but I've already noticed that some people feel quite sad and bitter about it.
I'd like to see more concrete examples on this. You might be right, though. Any project beyond a certain size has political issues to cope with (political: inter-personal issues).
 
The same as the other members of the BSD family (and as Illumos): each of them is a cocoon and touching any single thread of it, even to make it more comfy or look nicer, is considered a vital risk by its inhabitants. They just want it to stay the same until they die, full stop.
Yet FreeBSD has changed (major changes) over the years. And still people use it! Amazing!
People will complain, because people don't like change (it is said that the only change we like is the vacation, but only if we planned it ourself). However FreeBSD users (most of them anyway) are skilled, learned and smart; they are able to look past the changed things, try out a new version of FreeBSD (with major changes) learn how to use it and carry on.

Personal experiences of major changes: ports -> packages, zfs(), devfs(5) (anyone remembering how we used FreeBSD before devfs?) and probably a lot more that I have forgotten.
 
because people don't like change (it is said that the only change we like is the vacation, but only if we planned it ourself). However FreeBSD users (most of them anyway) are skilled, learned and smart; they are able to look past the changed things, try out a new version of FreeBSD (with major changes) learn how to use it and carry on.

The fact that the community has endured change and proven that we can adapt is good evidence that we don't simply "dislike change". The fact is that we are nervous of regressions and incorrect decisions.

Some of the FreeBSD foundation news articles strongly suggests that someone there has drank the "cloud cool-aid" in that they pine for Git services (lets face it, probably Microsoft GitHub) and migrated the IRC to consumer Discord (with a buggy bridge).

These don't affect us (until it is too late) of course but I am also not so sure that donations will be going to the correct cause. Instead they will end up just feeding proprietary companies masquerading as "the cloud".

I trust the FreeBSD developers to make the right decision. They have proven to be correct in the past. I just hope it is the developers who are still making these decisions.
 
Matter for... what? Opinions are not an exercise of power.

(also, really, "this isn't Linux", what kind of ad hominem is that?)
 
Opinions are not an exercise of power.
I'm saying that the FreeBSD people listen to users and don't have a Poettering or Linus, etc. But, really, that's like saying your elected leader listens to you when one needs to realize that political leader may have tens of thousands and more constituents. The best he can do is get a finger on the pulse, not listen to each opinion individually. Then again, how many "write your congressman" anymore?

Fun fact, it does work at times. Both of my sons are home schooled. One wanted to get into law enforcement and applied for training but was rejected by the state officials because he didn't have a high school diploma. However, he graduated summa cum laude from the same state university in Criminal Justice. No amount of writing and calling seemed to sway them so I called my Congressman. I didn't actually talk to him but the lady who answered the phone put me on a conference call with the state office that rejected him with the simple reasoning that, as she put it, "doesn't a university degree trump a high school diploma?" Needless to say, that went well.
 
Status
Not open for further replies.
Back
Top