[gentoo-dev] IRC cloak
Hi, I would like to use my developer IRC cloak but can't remember how to. Which got me to Google this question but so far I haven't come up with actual instructions for developers on how to do this sort of thing. Do we have any instructions? Could you point me to it? Or tell me how to use the cloak? Thanks already, Nick
[gentoo-dev] Last rites: dev-php/ffmpeg-php
# Brian Evans (09 Dec 2016) # Masked for removal, wrt bug 602164. # See https://github.com/PHP-FFMpeg/PHP-FFMpeg for a code based # replacement dev-php/ffmpeg-php signature.asc Description: OpenPGP digital signature
Re: [gentoo-dev] IRC cloak
On Fri, Dec 09, 2016 at 08:17:06AM -0700, Nicolas Bock wrote: > I would like to use my developer IRC cloak but can't remember how > to. Which got me to Google this question but so far I haven't come > up with actual instructions for developers on how to do this sort > of thing. Do we have any instructions? Could you point me to it? > Or tell me how to use the cloak? It looks like your IRC handle wasn't used enough and your nickserv registration expired. Re-register it, then talk to the freenode group contacts in #gentoo-groupcontacts. -- Robin Hugh Johnson Gentoo Linux: Dev, Infra Lead, Foundation Trustee & Treasurer E-Mail : robb...@gentoo.org GnuPG FP : 11ACBA4F 4778E3F6 E4EDF38E B27B944E 34884E85 GnuPG FP : 7D0B3CEB E9B85B1F 825BCECF EE05E6F6 A48F6136 signature.asc Description: Digital signature
Re: [gentoo-dev] IRC cloak
This is to prove my identity. My freenode nick is 'nicolasbock'. Thanks, Nick On Fri, Dec 09, 2016 at 04:38:20PM +, Robin H. Johnson wrote: > On Fri, Dec 09, 2016 at 08:17:06AM -0700, Nicolas Bock wrote: > > I would like to use my developer IRC cloak but can't remember how > > to. Which got me to Google this question but so far I haven't come > > up with actual instructions for developers on how to do this sort > > of thing. Do we have any instructions? Could you point me to it? > > Or tell me how to use the cloak? > It looks like your IRC handle wasn't used enough and your nickserv > registration expired. > > Re-register it, then talk to the freenode group contacts in > #gentoo-groupcontacts. > > -- > Robin Hugh Johnson > Gentoo Linux: Dev, Infra Lead, Foundation Trustee & Treasurer > E-Mail : robb...@gentoo.org > GnuPG FP : 11ACBA4F 4778E3F6 E4EDF38E B27B944E 34884E85 > GnuPG FP : 7D0B3CEB E9B85B1F 825BCECF EE05E6F6 A48F6136 signature.asc Description: PGP signature
[gentoo-dev] Re: Please stay on-topic.
On 12/08/2016 02:15 PM, Andreas K. Huettel wrote: -BEGIN PGP SIGNED MESSAGE- Hash: SHA512 Am Donnerstag, 8. Dezember 2016, 15:08:17 schrieb james: On 12/07/2016 04:39 PM, Robin H. Johnson wrote: On Wed, Dec 07, 2016 at 04:01:53PM -0500, james wrote: Can you cross post to gentoo-dev? I'm not subscribed to that list. Should not a wider community, particularly devs be part of the discussion? Please DO subscribe. Nope. I strongly believe that if your wider dev community had a deeper understanding of the responsibility chain... Even if I'm repeating myself... Please keep the mailing lists on topic. Problems caused by gentoo developer behaviors, are "on topic" for this list. Your position only serves to obfuscate remedies. While the topics of gentoo-project are widespread, conspiracy theories and US politics are definitely not on-topic here. And this sort of analysis is at the heart of why Gentoo is going to be categorized, as a "Boys Club" or a "tax dodge" or a "Fiefdom in the eyes of the IRS. The Gentoo Code of Conduct is quintessentially important to future viability of the distro. The devs all need to hear about these issues. Read on and you will see an action plan, the reasons for it and bit more detail, for those that require a bit more anecdotal evidence. Satisfactory resolution can not be achieved by the Council, nor the Foundation. It must be embraced by the wider gentoo dev community. In fact remedies proposed here can be implemented by a hand full of astute devs, should their convictions diverge from the council or foundation. Therefore, this is a paramount issue for those with dev status and those of us that have been participating with gentoo for quite a few years and have first hand witnessed these dev-induced malfeasant actions. *please listen* and we can save this distro from the fiefdom of constrained control, largely made possible by the dev community and the senior (privileged) devs whom also have a tainted history with Gentoo. Many senior devs have "clean hands" as we all know. Gentoo is not alone, there is, Alpine linux, now the fiefdom of Docker, and the linux kernel project are both viable candidates too, for scrutiny. In fact there are many charity-tech organizations (particularly 501(c) which can be portrayed as not being open but merely tax-avoidance schemes; this effects us all, deeply. Also, to confirm what robbat2 suggested, Gentoo Foundation organizational questions including how to deal with the IRS (should that even be discussed on a publicly archived list?) find their best audience on the nfp list. This can all be cleaned up. But decisive action needs to occur. Just read on or lodge your complaints with the Council if you feel the need. After all, whoever wants to participate can subscribe there, and whoever doesn't subscribe there probably doesn't *want* to hear about it. Where are you statistics? Let's hold a vote and ask gentoo-users to participate, if they like. A move to silence would be very interesting as a point of argument against your position. Perhaps Mr. Robbins can weigh on his experiences and perspectives? Other, bitter devs that have left? Countless others that have been mistreated (at least in their eyes) during their attempted journey to dev status? SO:: Only three things need to occur, to fix this mess. The past is not a problem, if when confronted with the truth the distro leaders take significant corrective actions. Converting what accounting records are known to gnucash, is trivial. So here are (3) actionable steps that can be achieved in short order and some of the reasons from the IRS point of view as to why they are of timely, actionable significance. (1) A GLEP that expedites and makes forking gentoo, relative as easy as possible. (2) An easy and straightforward method(s) to install gentoo. Stage-4, CD, ansible, ignition etc etc. (3) A documented pathway to become a dev on gentoo, with all requirements, tests and re-testing documented fairly and applied to all, new devs and existing devs alike. Why (1):: This allows most anyone be a gentoo dev. Lots of viable forks can be a home for learning, training and development of a wide diversity of different levels on technical competence, demonstrated by successfully being 'a gentoo or gentoo-fork developer'. Technologies like Java, could easily be supported should they choose to allow non-sources into the ebuild process, for their gentoo-forked-distro. Each fork can have their own dev rules. Anyone can fork gentoo and the master of that fork, manages that gentoo-sub-culture as they deem reasonable. Migration to dev-proper, can be a more difficult pathway, but needs to be open and fair. Training is conducted in the forks; one fork could be just for training for those seeking dev-status in gentoo-proper. Seriously, being a dev is no big deal. The quintessential quality or skill, is knowing which ebuilds
Re: [gentoo-dev] Re: Please stay on-topic.
On Friday, December 9, 2016 3:24:44 PM EST james wrote: > On 12/08/2016 02:15 PM, Andreas K. Huettel wrote: > > > > > Even if I'm repeating myself... Please keep the mailing lists on topic. > > Problems caused by gentoo developer behaviors, are "on topic" for this > list. Your position only serves to obfuscate remedies. Be careful, let what happened to me on -project stand as a warning. Short of this reply I am mostly radio silent, no posts, no bugs, no irc, nada... I was encouraged for such. Most do not care about the big picture, just their neck of the woods. Your posts make mine seem abridged :) A few select comments, as most rather not hear from me entirely. I will focus on technical ones. > Technologies > like Java, could easily be supported should they choose to allow > non-sources into the ebuild process, for their gentoo-forked-distro. This is already taking place in Gentoo. There is not the man power, time, or experience for larger Java applications to be packaged. They are going into tree as binaries. Most without even unbundling deps that are packaged in tree. > This enables (by active promotion) any individual, corporation or > business to benefit from a deployment of gentoo or a customized version > of gentoo, just like a privileged few corporations have benefited > (google, CoreOS etc). Without many companies at least testing and some > using gentoo, Gentoo is mostly a fiefdom of certain companies. This is > evidenced by the history of devs that work at these companies and have > been or currently are gentoo devs. It's back-channel control, evidenced > by irc and other venues where only certain folks can listen. This is the down side of letting everyone work on something that pertains to them. Without any sort of over all leadership, a common goal or direction from the Council. I am not sure Gentoo has had such leadership since the Chief Architect position was eliminated. It is one thing for a developer to work in an area, scratch their itch. But as companies hire developers. Those companies can start to take things in their own direction. Largely this has not been a problem, but it lets outside companies indirectly control the direction of Gentoo. If a conflict between companies arises, it could get interesting. Thankfully not an issue and may never become one. > WHY (3):: Update the exam every (2) years and require all devs take the > exam to re-qualify. Interesting comment as someone who has done the quizzes several times. Most developers do them once and never again. I wonder how many would pass a quiz if they were retested Also something many here are likely not aware. To become a Gentoo Java Developer, there are 3 quizzes. There is a 3rd Gentoo Java Quiz. I am not aware of any other quizzes for other parts of the tree. Which in part goes to show some of the complexities with Java. Though it may be beneficial if others come up with their own quizzes, which could be small. There are some nuances to packaging other languages. Not sure if as many as Java to require another quiz, but may help for others joining teams for other languages. https://wiki.gentoo.org/wiki/Project:Java/Developer_Quiz -- William L. Thomson Jr. signature.asc Description: This is a digitally signed message part.
Re: [gentoo-dev] Re: Please stay on-topic.
On 12/09/2016 03:58 PM, William L. Thomson Jr. wrote: On Friday, December 9, 2016 3:24:44 PM EST james wrote: On 12/08/2016 02:15 PM, Andreas K. Huettel wrote: Even if I'm repeating myself... Please keep the mailing lists on topic. Problems caused by gentoo developer behaviors, are "on topic" for this list. Your position only serves to obfuscate remedies. Be careful, let what happened to me on -project stand as a warning. Short of this reply I am mostly radio silent, no posts, no bugs, no irc, nada... I was encouraged for such. Most do not care about the big picture, just their neck of the woods. Your posts make mine seem abridged :) Hey very cool advise. But I'm an old maverick. Hell I was 86'd from gentoo user at least once, probably twice for offering to pay devs to create ebuilds, that my small company could use and would be freely available to all. Somehow google does the same thing, except it's a regular paycheck and that is OK. I ask to do it and get banished for a while. Fiefdom. A few select comments, as most rather not hear from me entirely. I will focus on technical ones. Technologies like Java, could easily be supported should they choose to allow non-sources into the ebuild process, for their gentoo-forked-distro. This is already taking place in Gentoo. There is not the man power, time, or experience for larger Java applications to be packaged. They are going into tree as binaries. Most without even unbundling deps that are packaged in tree. Yep. Apache Spark is one of my java complaints (I mean bugs) BGO-523412. There is a purported version of Apache spark, written in Python; "dpark" I believe is the name, not sure it's still active. But I'm not allowed to pass out beer money, or a weekend in Vegas money for things that benefit the gentoo community, or so I'm told Apache-Spark the blocker and is a really cool package that can be added to Apache-Mesos, for cluster acceleration. But Apache Mesos is a competitor to other favored cluster codes at Gentoo. Fiefdom? This enables (by active promotion) any individual, corporation or business to benefit from a deployment of gentoo or a customized version of gentoo, just like a privileged few corporations have benefited (google, CoreOS etc). Without many companies at least testing and some using gentoo, Gentoo is mostly a fiefdom of certain companies. This is evidenced by the history of devs that work at these companies and have been or currently are gentoo devs. It's back-channel control, evidenced by irc and other venues where only certain folks can listen. This is the down side of letting everyone work on something that pertains to them. Without any sort of over all leadership, a common goal or direction from the Council. I am not sure Gentoo has had such leadership since the Chief Architect position was eliminated. It is one thing for a developer to work in an area, scratch their itch. But as companies hire developers. Those companies can start to take things in their own direction. Largely this has not been a problem, but it lets outside companies indirectly control the direction of Gentoo. If a conflict between companies arises, it could get interesting. Thankfully not an issue and may never become one. I have no problem with this. I have a large problem with gentoo distro management, effectively via policies, constraining commerce. Sorry, I'm not one to snitch, however when the ICEMAN comes calling, I am going to laugh out loud and share the pain with others. It's been a long time coming. definitely Fiefdom. WHY (3):: Update the exam every (2) years and require all devs take the exam to re-qualify. Interesting comment as someone who has done the quizzes several times. Most developers do them once and never again. I wonder how many would pass a quiz if they were retested Are you espousing principals of truth, equity and otherwise, "do unto other as you'd have them do unto you"? I believe that's how Pense rolls, from what I read but some lawyer buddies from Ohio say he is the ICEMAN. Also something many here are likely not aware. To become a Gentoo Java Developer, there are 3 quizzes. There is a 3rd Gentoo Java Quiz. I am not aware of any other quizzes for other parts of the tree. Which in part goes to show some of the complexities with Java. Been down that path. The exam was not ready a few years ago. I've moved on. But I'm just a java hack because the way things are done, it's just a voluminous source of ever changing codes and new patches that requires a full time attention to be effective. By the time I hack something java, it the old unsupported way to do things. I'm fine with using canned modules and binaries, I just think gentoo ought offer a secure sandbox, VM, container or whatever for java. I do understand the concerns so it is frustrating and I just do not have the time to become a java whiz, unless writing your own rxtx counts? I built a serial data analyz
Re: [gentoo-dev] Re: Please stay on-topic.
Staying technical, would say more, but I am already saying more than anyone would want me to, since most want me to go away... On Friday, December 9, 2016 5:21:53 PM EST james wrote: > > Been down that path. The exam was not ready a few years ago. I've moved > on. But I'm just a java hack because the way things are done, it's just > a voluminous source of ever changing codes and new patches that requires > a full time attention to be effective. By the time I hack something > java, it the old unsupported way to do things. While that literally was the case when I joined, 2006, when generation 2 Gentoo Java eclasses and such were being hashed out. I would revise an ebuild, commit, things would change, revise, commit, repeat... However things have been pretty stagnant on the Java front for some time. Very little eclass changes. I have made effort to document the eclasses and Java ways of doing things on Gentoo. Having helped to make the quiz years ago which also could use an update. Though since I have some disagreements with devs who from time to time who poke Java stuff. I have pretty much stopped. With recent actions taken against me on -project, I doubt I will ever resume. https://wiki.gentoo.org/index.php?title=Java_Developer_Guide&action=history https://wiki.gentoo.org/index.php? title=Gentoo_Java_Packing_Policy&action=history > I'm fine with using > canned modules and binaries, I just think gentoo ought offer a secure > sandbox, VM, container or whatever for java. I do understand the > concerns so it is frustrating and I just do not have the time to become > a java whiz, unless writing your own rxtx counts? I built a serial data > analyzer that sniffs the physical serial interface too, just for > grins.. actually for a customer. Java on Gentoo is really not bad, if you are familiar with Java at all. -- William L. Thomson Jr. signature.asc Description: This is a digitally signed message part.
Re: [gentoo-dev] Re: Please stay on-topic.
On Fri, Dec 9, 2016 at 4:36 PM, William L. Thomson Jr. wrote: > Java on Gentoo is really not bad, if you are familiar with Java at all. > It's actually quite insane "if you're familiar with Java at all"... Building C and C++ from source is great. 51% of dev-java category is doing pointless work. 401 ebuilds only have IUSE="doc source" which can almost always be fetched from Maven Central, and 44 ebuilds have no USE flags at all. That's just from simple grep results. Given the ugly majority there, I don't doubt there's some silliness going on in the remaining 49%. Building Java from source to get the exact same jar file every time on a million machines when you could just fetch the upstream jar instead is plain stupid.
Re: [gentoo-dev] Re: Please stay on-topic.
On Friday, December 9, 2016 5:06:14 PM EST Gordon Pettey wrote: > On Fri, Dec 9, 2016 at 4:36 PM, William L. Thomson Jr. > > wrote: > > Java on Gentoo is really not bad, if you are familiar with Java at all. > > It's actually quite insane "if you're familiar with Java at all"... I have been coding in Java since 1.3, going back to 2001 I think. > Building C and C++ from source is great. 51% of dev-java category is doing > pointless work. In what sense? What source/target are pre-compiled jars? That same thing could be said for perl, python, ruby, and likely others on Gentoo. However they tend to be the same, they have no source/target like Java. > 401 ebuilds only have IUSE="doc source" which can almost > always be fetched from Maven Central Fetched in another step, this allows it to exist locally. But many are moving to Gradle. Maven is a bit old school now, one step beyond ant. > , and 44 ebuilds have no USE flags at all. They may be upstream -bins, Sun had a fair amount with no source release. Could be the doc/source were omitted on accident or purpose. > That's just from simple grep results. Given the ugly majority there, I > don't doubt there's some silliness going on in the remaining 49%. Building > Java from source to get the exact same jar file every time on a million > machines when you could just fetch the upstream jar instead is plain stupid. Really is it the exact same jar? You know what changing source/target does right?. Which Gentoo does have some issues there, as it does not use older rt.jar when using targets < current JVM version. https://blogs.oracle.com/darcy/entry/bootclasspath_older_source Not all jars are the same, not all Java binaries are the same. Tossing a 1.7 jar into a 1.8 JVM does not really give you any 1.8 benefits. It will run as if it was 1.7 in a 1.8 JVM for example. -- William L. Thomson Jr. signature.asc Description: This is a digitally signed message part.
Re: [gentoo-dev] RFC: Proposal for addition of distribution variables
-BEGIN PGP SIGNED MESSAGE- Hash: SHA256 On 04/12/16 21:28, Michael Orlitzky wrote: > Maybe a GLEP with the contents of your first mail? If there's > consensus for the premise and the implementation, it's easy to add > the variables. Then it's only a matter of updating ebuilds. I've tweaked some of the wording and changed to Wiki format from rST to conform to GLEP 2, and filed bug #602202 [1] to have this formally added as a GLEP. Comments are definitely still welcome, either to this list/thread or my personal email. I didn't find any reference in GLEP 1 to how changes can be made to GLEPs already posted on Bugzilla, but I am definitely open to hearing any that the community has to suggest. - --arw [1]: Bug 602202: https://bugs.gentoo.org/602202 - -- A. Wilcox (awilfox) Project Lead, Adélie Linux http://adelielinux.org -BEGIN PGP SIGNATURE- Version: GnuPG v2 iQIcBAEBCAAGBQJYS1jDAAoJEMspy1GSK50UXN8P/09uND1/3WMFDT/sn+/gvooR nFd0V3FHmljoy4EOOTUceYQQBtAPuzRLy8rfNIOtXGyxfx/oHrH1zgqrfugQwl86 KwuXh5t4v5dUQB12TjKtTItITWx48KoedcXM4Pxu73nKo1L7r/jySFFAgOvqDf/P vdrGOsa1R4uyg/XvC8fDojL5m9SAF60vloHUhQChIp9vhTzxs8PRVi8gIffuZGTq zrHczroutx4FA0LdsPosEQs9ClO3rbBk/JTuG6WAkgqTeMWDXXAxOyNYMdiQworA cuO0eU0S3pJcIE6NsbKmZ0qbcJbsgekv/UyBy08r06maOnEt3LbzPRVcOSc50qN8 qQ39DPHnda1bQXINwzeTKq7/2Oly9TjJi4P9c4mUNTeAayRwcvfo/8G16E96/eyj z2YHYbJytl945S9B1W7fGCzEuCEVOZiXXBvMpu1bO+B0Oonp3zOhwKgqBE/dRZvl 8dAGQzA7LMXb2i1OUTUdtTdbY3cju0t+yYITJ64LKdLWYWgY/m90vttkYE7CHyJO U33pShFVQuVkHjccT+oWCtI9anZ6Lnjr75PKbi7GlMihQ2hocxH6dmnUZksquLcx HWE1OZSESVqBUH5PHlBQuJsjVwkrKEdv9Yqtlr0mfk7BYFw9r/5b//pnXzjCqO5e y+zanHProXl06FjtA9nd =1NJI -END PGP SIGNATURE-
Re: [gentoo-dev] Re: Please stay on-topic.
Hi, On Fri, 9 Dec 2016 15:24:44 -0500 james wrote: > On 12/08/2016 02:15 PM, Andreas K. Huettel wrote: > > -BEGIN PGP SIGNED MESSAGE- > > Hash: SHA512 > > > > Am Donnerstag, 8. Dezember 2016, 15:08:17 schrieb james: > >> On 12/07/2016 04:39 PM, Robin H. Johnson wrote: > >>> On Wed, Dec 07, 2016 at 04:01:53PM -0500, james wrote: > Can you cross post to gentoo-dev? I'm not subscribed to that list. > Should not a wider community, particularly devs be part of the > discussion? > >>> > >>> Please DO subscribe. > >> > >> Nope. I strongly believe that if your wider dev community had a deeper > >> understanding of the responsibility chain... > > > > > > Even if I'm repeating myself... Please keep the mailing lists on topic. > > Problems caused by gentoo developer behaviors, are "on topic" for this > list. No, they are not. As I remember from my quizzer gentoo-dev is intended for technical discussions only. For legal stuff we have nfp mail list, for general project related stuff which doesn't fit cases above we have project mail list. The reason we have such separation is that different people have different interests, e.g. some developers are not even members of the Foundation, because they have no interest in legal stuff and it is their right to be so. One of the reasons we have separation between the Council and the Trustees is to relieve developers not interested in legal stuff from legal stuff. And now you are proposing to discuss legal matters on gentoo-dev. This is terribly wrong. Is it that hard to subscribe on gentoo-nfp mail list? It is open for everyone and is created for exactly these kind of discussions: legal and other Foundation-related stuff. Best regards, Andrew Savchenko pgpcW8c_FlM4X.pgp Description: PGP signature
Re: [gentoo-dev] Re: Cross Post due to technical component - Thanks for all the fish
On Wed, 7 Dec 2016 12:15:06 -0500 james wrote: > Being able to use stage-4 or stage-5 (G. forums) installs to rapidly > provision a collection of bare-metal systems [BGO-593218] into a wide > variety of hardened clusters is my passion. Unikernels as stage 4 > packages can then very easily be targeted for very specific needs: VM > or container or bare-metal. Gentoo-proper is has too much political > baggage to encourage folks to innovate, imho. So, I really hope the > gentoo dev community gets behind the Anna Wilcox idea of streamlining > Gentoo into the most fork-able distro on the planet. WE could all be > one happy family and yet be very competitive with our ideas, trials > and published results? Surely a few eggheads (academcis/pedantics) > see the wisdom of competing micro_distros? Not unlike competing > micro_breweries, it make the entire craft much stronger and better > for all. > > > Then there can be peace and harmony as everybody can do exactly as > they please with their little cluster of gentoo and their very own > portage-tree. And then folks running gentoo-proper now can pick and > choose which innovations they want to include in the master tree. > Isn't that pretty much what Google and CoreOS do now, as well as the > gentoo derivative OS? Why not accelerate what has worked, for the > few, to emancipate those of us still chained into user-land servitude. As an ordinary user, this sounds pretty bad. Forking is great for developers, but bad for users. I don’t *want* 27 different Gentoo-derived fork distributions, each of which is great at one thing. I don’t want to have to reinstall a different OS just because I switch from writing embedded code to running Octave. Honestly, I don’t even want to go out and find other OS’s repos, add them as overlays, and hope the inter-OS dependencies work. As an ordinary user, what I *want*, is to install one OS and not think about it again. Ideally, Gentoo. When I want to do embedded development, I just emerge dev-embedded/thingy. When I want to do some math, I just emerge sci-mathematics/octave. Most things that most people care about in the main tree. Breaking things up into overlays or different OSs or whatever just means adding more hoops that I have to jump through before I can start working on a new topic. -- Christopher Head pgpa8a3bVCWvb.pgp Description: OpenPGP digital signature
Re: [gentoo-dev] Re: Cross Post due to technical component - Thanks for all the fish
-BEGIN PGP SIGNED MESSAGE- Hash: SHA256 On 09/12/16 23:46, Christopher Head wrote: > On Wed, 7 Dec 2016 12:15:06 -0500 james > wrote: > >> Gentoo-proper is has too much political baggage to encourage >> folks to innovate, imho. So, I really hope the gentoo dev >> community gets behind the Anna Wilcox idea of streamlining Gentoo >> into the most fork-able distro on the planet. WE could all be one >> happy family and yet be very competitive with our ideas, trials >> and published results? Surely a few eggheads >> (academcis/pedantics) see the wisdom of competing micro_distros? >> Then there can be peace and harmony as everybody can do exactly >> as they please with their little cluster of gentoo and their very >> own portage-tree. And then folks running gentoo-proper now can >> pick and choose which innovations they want to include in the >> master tree. > > As an ordinary user, this sounds pretty bad. Forking is great for > developers, but bad for users. I don’t *want* 27 different > Gentoo-derived fork distributions, each of which is great at one > thing. I don’t want to have to reinstall a different OS just > because I switch from writing embedded code to running Octave. > Honestly, I don’t even want to go out and find other OS’s repos, > add them as overlays, and hope the inter-OS dependencies work. I think James has perhaps spoken ambiguously, or at least I hope that you have misunderstood his proposal. (If you haven't, then he's misunderstood mine.) The point of making it easier to fork is not only for the benefit of developers. As James says: > And then folks running gentoo-proper now can pick and choose which > innovations they want to include in the master tree. The idea being the people who "run" Gentoo, that being the developers of Gentoo, can pick what they want from the forks and derivatives, and include those improvements in the master tree. Then all Gentoo users, and all derivatives of Gentoo, can benefit from those improvements. Consider the relationship between Fedora and CentOS/RHEL. Fedora is released rapidly, compared to RHEL. It is where innovation and development happen for them. Then RHEL picks the best bits from them and ships it in their product. You don't have to run Fedora to be able to use the work they produce. (Though sometimes you have to wait a while!) So for one example, at Adélie we are focusing hard on the musl libc. At some point in the future, when we have things looking good, we can contribute that back to the official Gentoo musl overlay. Ideally, that would be the main Gentoo package tree... but at least the overlay. We have also packaged some great open fonts that we've found. We can easily send our ebuilds to Gentoo's media team and they could put it right in to the tree. (Right now, I'm still working out the best ways to use the fonts eclass... hence there is no upstreaming yet.) Forks and derivatives allow a much wider community the ability to experiment with the powerful Gentoo system without fear of "breaking" the "real" Gentoo tree. Things like my APK BINPKG_FORMAT patch may never make it upstream, which is fine. However, overall the goal is to enrich the broader Gentoo userbase. After all, isn't that the idea behind open source in the first place? You have the freedom to take the code, do what you want with it, and then contribute your changes back when you're sure they're good. Forking Gentoo allows people to try out more wide-sweeping or drastic changes without any danger. The future can be cool and groovy if we have the freedom to tinker :) - --arw - -- A. Wilcox (awilfox) Project Lead, Adélie Linux http://adelielinux.org -BEGIN PGP SIGNATURE- Version: GnuPG v2 iQIcBAEBCAAGBQJYS5zfAAoJEMspy1GSK50UjiYQALxqN9b3UG04ioErJ/fyBoaK qSZjyCw7xXK+SaiNXyfDQPPmoTMxdNgog74awEwM4bGVYplMECeIf8JLcyDpRzol fBbnhucckeLAYM+n4RNv/eozjRtg7qc5SgnkIL0mihDkzEVgAX5d5pUS4V4ZIoe5 P8Q3fMsxdOFomBetLG3pKBpO980xylf2xy/6EoZVAbeR0kIqw4NecskTe+by4toz vJbrvKX4ht+yhNPGw+QfKY+oM3KEzc8VsjcDI53OzFL4CuNm43CkAECExhcl1pXi 4VbmENP5M1omP5AhAJiEsev3ORhzXKFX+9Zs8Z/WQYi+Osnzw3I2HxX5FK7g8J9K DNprGIrjnoazwKVMaBapK8qEmI8r8xYQVqKq6s8wzWbTa8k1FYA0H8A/pCbeQmjz o/TdE8oc5py426T7CThxFVsRdLiq0q8werEJ4Zql1nFBNYu34Us15i8MIkujHu25 mrByesaeuTM25TfHzRV0A7LCte8vvGJkwZ6Z9ndokJdSIn9Xjw4sUgGRjT5SKsu4 KKN4UDTpATSX5jRmCfVeREHWyPuVJermeX/2BRVmH1EbQ4KgqPetLMm19SBzKxEs dOLLlPRj4lsh9s7Z/J9nkzKUGWsNBUGbM9+iMOF5/e8CgT4eLIcmHsmFeqxJsylk IrzjcKTPHvEeM4oP+Yfm =2Y0G -END PGP SIGNATURE-