Rich Freeman <rich0 <at> gentoo.org> writes:
> On Fri, Jul 24, 2015 at 8:56 AM, James <wireless <at> tampabay.rr.com> wrote: > > Rich Freeman <rich0 <at> gentoo.org> writes: > > > >> Just be practical. From my experience showing up at a LUG and telling > >> 20 > >> people how something worked well for you gets you a lot further than > >> handing out free T-shirts and hats at a booth. > > > > Rich, I'll be practical. Gentoo needs an installer program, like most > > other distros if you want your rank_n_file users to entice new users. > Somehow we got you to use it. :) Not really. A bsd hack I know introduced me to gentoo. He has since moved back to bsd over the systemd flack.... He developed a bsd ramdisk for tor quite a few years ago..... I told him that gentoo (glep 64) is working on a DAG so all those little nuisance files can be tracked and itemized. He just laughed and laughed and laughed. What does this mean? Don't read too much into it. That's my background on joining up with Gentoo. He just like the BSDish ports setup. YMMV. > If somebody wants a distro like most other distros, they should just > use one of those other distros. I don't get paid by the Gentoo > install. I want people to use Gentoo because it is the right solution > for them, not because it makes me look more important. That's damaged logic. If somebody (many) want an installer then they have to want a binary distro? Sorry, your logic is flawed linking those 2 issues. The nature of the installation does not limit one to a binary or rolling distro; separate issues. > But, as has been pointed out there is already an installer project, > and anybody is welcome to create as many different installers for > Gentoo as they like. Yep that's the point exactly. A well formulated installer will allow many folks to build upon that offering, as they like. > > Second, you need something to attract a few local sponsors to have > > a viable > > LUG, imho. Something they cannot get elsewhere from other distros. > I attend a LUG monthly and the only sponsorship it gets is the > building space it meets in - and it doesn't have trouble finding them. So many LUGs died in many cities because anyone can download an installer and put buntu or whatever distro on a box themselves. Now building a cluster, running btrfs in raid1, yubikey setup and usage, encrypted file systems using TOR and such as those advance features are what would attract folks back to a LUG, imho. YMMV. Here in the Tampabay area, >3MM folks, there's not a single, active LUG. > And LUG stands for "Linux User Group" not "Gentoo User Group." I > wasn't suggesting that we should be going out and creating LUGs > (though if an area lacks one that would probably be a good idea), but > rather just talking about Gentoo at whatever LUG you would otherwise > attend. I talk about Gentoo all the time. I'll probably be > co-presenting with an Arch developer at a future one, and a few > attendees have used Gentoo. Most have favorable impressions of > Gentoo, though many don't really think it is for them, and that is > fine. I'm certain to point out that the Chromebooks in the room are > based on Gentoo, and I've seen presenters at the LUG point out the > same even when they don't use Gentoo themselves. That's the > difference between education and propaganda - people who work in such > circles tend to recognize the value of the former. At local > conferences RedHat can hand out bottle opener USB sticks and people > smile and appreciate them, but it isn't like they go wipe their > systems and install Fedora on them when they leave. And plenty in the > group do use RHEL/CentOS/Fedora - and appreciate when it is and isn't > the right tool for the job. > > I see one of our devs just updated openstack:: [1]. What I did not see > > is the corresponding (probably more important) optimized kernel config > > and the kernel/system profiling wiki on how to tune servers for > > clusters. > Well, go write one. :) Exactly. That's what you do at a LUG, imho. Show what buntu users cannot do themselves..... (again ymmv). > It doesn't make sense for us to tell the guy updating openstack that > he isn't allowed to do it unless he also updates the wiki and provides > configurations for tuning clusters. It also doesn't make sense to > forbid somebody from authoring an article on tuning clusters unless > they commit to update it the day any openstack update is done. Your drawing way too many conclusions. There is nothing intended like that in what I said. I was merely pointing out that work builds on other work and it takes a community. You cannot install a cluster, if one is struggling with an install. Note only that many folks do not understand that a cluster needs a specialized, (tuned if you like) kernel. I'm scratching my head wonder how you jump to these conclusion. I do not expect everyone to agree with me. Clearly there are multiple positions on installers for gentoo; what clustered software to use and how to customize the servers and other related issues. Stating things does not mean I am condemning anyone for what they contribute; please stop reaching into what I write as being deleterious. I do not have that intention. But how to get an installation useful, like customizing the kernel underneath, or pursuing RDMA (RoCE) on the underlying file system, or any number of related issues I have posted about ARE related to the work of clusters. I'm happy folks are working on openstack. I like mesos more, but both are very wonderful. Either way, kernel tuning and systems performance profiling are also critical components. That's it. No condemnation, just pointing out the obvious, so others with less cluster experienced folks know what they have to deal with. How you get "he isn't allowed to" is non sequitur as an argument. > So, people are free to work on some things and not others, and if you > want to see the other things happen you need to step up and offer to > do it. I have several gentoo projects. Progress is slow, as you know. I do my part. > > YOU or the other leaders want lots of new members? It's a rhetorical question; surely you know about these? Yes, we all want new users to keep the distro healthy; a few will be like Zac. I remember when he first showed on the gentoo user group. Fresh ideas also arrive via new members, and that's a very good idea. > If you're waiting for somebody else to make Gentoo into what you want > it to be, you may be waiting for a long time. Gentoo is what it is > because of people who stepped up and contributed, and did more than > post on lists. By all means offer suggestions, but taking an > aggressive tone with those who are contributing is one of the reasons > that few of the developers actually read this list, which is a pretty > sorry state of affairs. Aggressive? Come on. A strong, well argued opinion is hardly an aggressive tone. It's a discussion thread; so posting or replying is evidence that folks have options. Just because they are 'strong opinions' does not detract from the validity of the need. After all Maffblaster took this need up on his own, not at my be-questing. Sure I participated in the discussion but aggressively? Rudely? > However, people who contribute aren't obligated to read your complaints, > and I'm certainly not going to be the one to change that. So? Arguments as to why an installer is needed is now my complaints, in the context of the installer? Public relations? Surely you see that an installer or lack thereof is pretty much the face of gentoo's public relations, or are you not willing to read that into your arguments like you extended my arguments for an installer? For the majority of folks new to gentoo, the installation process is what they see. That installation experience is the main Gentoo PR statement. > Gentoo is what YOU make of it. That's how it got to where it is > today. That's what makes it so great. It isn't shaped by some > corporate vision. There isn't some guy with lots of money in the BDFL > role. If nobody in the community wants a feature there is no paid > developer team who is going to ram it down your throats anyway. The > things that make Gentoo what it is are there because somebody in the > community cared to create them, and maintain them. > The role of the leadership is to foster that environment. It is to > create an organization where contributions are welcome. Really? Then why are there so many posts against an installer, mostly from folks who have been around this distro for a long time. Maffblaster is willing to take on this project and look at all of the negative flack for his idea to contribute what he thinks gentoo needs. > It is to > create an atmosphere where the value of contributions is recognized > even if we don't personally find them useful. It is to create a place > where with different priorities and preferences can still collaborate > without creating undue burden on anybody to support something they > don't care for. We can easily have a separate group for install help; or a new group for expert users. It's a non issue. I'm just surprised how many folks are against an installer, when all they have to do is ignore it's presence. Linking of an installer to a binary distro is not logical (non_sequitor mathematically). > And that is why if you create an installer, or a website promoting > easy ways to do things with Gentoo using an installer, I'll be happy > to be the first to thank you for it. Personally I'd rather see you > work on an easy-to-follow guide for setting up and managing Gentoo > systemd nspawn containers using Ansible, but who am I to tell you what > you ought to be spending your volunteer time working on? Alan is the ansible wiz.....? Nobody has been telling anyone what to do. It's been a discussion on the need or feelings about a new installer? Your reaching again into areas that are unrelated to whether the distro needs an installer. > That's basically the Gentoo way. Write code, not drama. But, if > you'd rather do other things for Gentoo than write code then do that, > not drama. :) You are *reaching* way to deeply, off topic here.... You, like others think that an installer, especially a graphical installer, is a bad idea, right? > > So what do you say? Do you really want to put Gentoo atop the > > distro-watch > > lists; again? (clustering and embedded are sexy;imho::ymmv). > Heh, if Gentoo cared about distrowatch we'd probably stick Gentoo > somewhere in our user agent strings for all our browsers. I look at > it maybe once a year at most, and when I do it doesn't count as a > Gentoo visitor. Most of the browser maintainers don't do that for the > same reason that we don't go changing default search engines or > spamming referrer IDs into all your GET/POSTs. As I previously stated 'sexy' means folks want to include you in their strategic plans. I never mentioned anything about web broswers... so that's your idea. > Distrowatch rankings just tell you what user agent strings are sent to > them when people visit the site. That is about as useful a metric as > asking who is paying the domain parking services to advertise their > webserver. An installer will allow more folks to test gentoo. Surely some (maybe a few maybe many) will keep using gentoo. I never went into a thesis on 'distrowatch'; again that's your exploration. All I was talking about was public relations via the new installer and some things that could occur. I could not care less about distrowatch; it was a generic reference only. Sorry if I offended you. But, you accepted council responsibilities. You can easily do what the other council members do (hide). If you want to discuss the merits of an installer, fine. But, I find it, as many do, really, really strange that gentoo does not have an installer. In fact one poster reminded us of a former gentoo dev that wrote the original installer that many distros now use, granted it has been enhance. Conclusively, from what you and many others have posted is there is a hatred toward what other do with their own time, like write a new installer; or have I done what you have and 'read' or 'reached' too much into your arguments? sincerely, James
