On Thu, Dec 12, 2013 at 8:01 PM, Otis Gospodnetic < [email protected]> wrote:
> Hi, > > Thanks David & Rainer. The last few emails were very constructive and > pragmatic. Based on what I heard I'd think you'd want to push things > towards Github. Ignore Apache for now at least. > * Make Github *the only* source (i.e. not Adiscon git) > why? > * Make > > https://github.com/rgerhards/rsyslog/blob/master/doc/build_from_repo.htmlmore > prominent (and markup, not HTML) > "markup, not HTML" easy said... > * Link it from rsyslog.com. Your github repo clearly doesn't have much > Google juice. Maybe David can pull some strings? ;) > > that'll happen tomorrow Thanks for being open. > > what saddens me is that we seem to be conceived as "closed". The root of rsyslog is that we wanted to provide a real open soure alternative to an (at that time) becoming more and more closed syslog-ng: http://blog.gerhards.net/2007/08/why-does-world-need-another-syslogd.html One success I had with rsyslog is that Balabit changed it policy considerably. Keep on mind that Adiscon is a small team of mostly peers, in contrast to a multi-million $ company... Rainer > Otis > -- > Performance Monitoring * Log Analytics * Search Analytics > Solr & Elasticsearch Support * http://sematext.com/ > > > On Thu, Dec 12, 2013 at 1:52 PM, David Lang <[email protected]> wrote: > > > On Thu, 12 Dec 2013, Otis Gospodnetic wrote: > > > > Hi, > >> > >> On Thu, Dec 12, 2013 at 12:18 PM, David Lang <[email protected]> wrote: > >> > >> how is an Apache project more "open" than rsyslog currently is? > >>> > >>> > >> I have limited Rsyslog "feel". But it seems there is a single company > >> behind it and the 24h problem was mentioned more times on the list in > the > >> last week than Kafka, so that says something ;) > >> > > > > well, comments of the 24 hour problem happen any time there is a to-do > > list and people ask why things aren't done. a project can have 1000 > active > > developers and still have a 24h problem. A higher user:developer ratio > > makes this mroe obvious, but it's not clear how chaning to Apache would > > change this. > > > > > > Rsyslog is probably not getting a huge number of patches on a regular > >> basis > >> and that is why Rainer can keep up. If that were to change patches > would > >> grow stale and contributor would be put off. This won't happen any time > >> soon, I think, because I suspect growth is limited with the current > >> development model. But that limited growth is what Rainer maaaaaaaay > want > >> to change. Or not - I don't know, but would love to find out, that's > why > >> I > >> asked the question. > >> > > > > well, if we started getting a lot of patches, we would probably get > > several contributers who are doing a lot of patches, and the kernel model > > of 'trusted lieutenants' would probably develop. particular contributers > > would work in particular areas, and they would review patches for those > > areas and send them up to Rainer for final acceptance. > > > > This scales very well and can address this limit. > > > > But since we are so far from this limit, it doesn't make sense to make > > major changes and loose months of development work when this isn't a > > problem now or in the reasonable future. There are a huge number of > > potential problems that could happen in the future. if we tried to > prevent > > all of them now, development would stop for years and 99%+ of the > problems > > would never have happened anyway. Wait to solve a problem until it's at > > least near. > > > > Rainer would love to get more contributions from people. > > > > > > Is it because it would allow companies to make proprietary forks of > >> rsyslog > >> > >>> and not contribute back? (something that currently they need to get a > >>> license from Adiscon to do) > >>> > >>> > >> Hm, no, I don't think this is a problem to be concerned with at all. In > >> the long run it pays to contribute back. Not very many people want to > >> fork > >> a healthy project. > >> > >> or is there some other benefit? > >> > >>> > >>> > >> See the mention of growth above. But if growth, more contributors, even > >> more corporate backing (i.e. company X paying its employees to work on > >> Rsyslog because their product relies on Rsyslog) is not interesting, > then > >> maybe the current model is OK. > >> > > > > moving a project to apache doesn't magically make it grow. It only makes > > it grow if it fixes some problem that was preventing growth. > > > > A log of companies have made projects 'open' on the promise that it would > > get them lots of new developers for no extra effort, and it almost never > > ends up working that way. > > > > This sort of thing is why I am hammering on details of what is expected > to > > change. Generalities don't convince me. > > > > > > > >> Rainer and Adiscon have been very open in terms of taking suggestions > >>> from > >>> outside and implementing sponsored improvements when they decided that > >>> the > >>> suggestions were too large a chunk to fit in the time budget for free. > So > >>> unless you expect more people to be able to tell Rainer what to work > on, > >>> the development direction would not be any more open than it is now. > >>> > >>> > >> Aha, I think this points out something important. > >> Suggestions are good. > >> Contributions are better. > >> Being able to suggest + contribute + commit - priceless. > >> > > > > Actually, I disagree. too many peopel with commit privileges are a > > problem, see the wikipedia dueling edits as an example. > > > > Multiple committers only work when they are in very close agreement > almost > > all the time. > > > > > > And it's not like with being open the business of "sponsored features" > >> goes > >> away! There are *tons* of companies whose business model is that + > >> training + support + .... while the code is all fully open, has a *big* > >> community around it, etc. etc. > >> > >> Openness would help with precisely what you outlined here. It's not > about > >> "tell Rainer what to work on". It's about "I'll do it, including > testing > >> and documenting and committing and later answering community's questions > >> about it and becoming an expert and writing a book about it and thus > >> giving > >> Rsyslog even more exposure and ...." you see how it goes, it's a chain > >> reaction. > >> > > > > except for the commit portion, everything you say is already possible, > adn > > there are enough very good projects where people do everything else you > > talk about without the ability to directly commit themselves that I > > disagree with the idea that commit privileges are critical. > > > > David Lang > > > > > > Otis > >> -- > >> Performance Monitoring * Log Analytics * Search Analytics > >> Solr & Elasticsearch Support * http://sematext.com/ > >> > >> > >> > >> > >>> > >>> David Lang > >>> > >>> On Thu, 12 Dec 2013, Otis Gospodnetic wrote: > >>> > >>> Date: Thu, 12 Dec 2013 12:09:50 -0500 > >>> > >>>> > >>>> From: Otis Gospodnetic <[email protected]> > >>>> Reply-To: rsyslog-users <[email protected]> > >>>> To: rsyslog-users <[email protected]> > >>>> Subject: Re: [rsyslog] Rsyslog => Apache? > >>>> > >>>> > >>>> It does mean that. Anything that is pluggable and that is GPL and > that > >>>> doesn't have to be in the Apache repo and could be pulled in at "make > >>>> time" > >>>> could stay under any other license. I've seen this done before. > >>>> > >>>> Re 2-3 months of work. One way to look at this is that by opening up > >>>> Rsyslog more you'd get more contributions (this *does* take time and > >>>> nurturing effort, but it's a win in the long term). If I were you, I > >>>> would > >>>> specifically "advertise" pieces that are under GPL and that are top > >>>> priority to rewrite. Not by opening a hard to spot issue, but in some > >>>> much > >>>> more prominent fashion. > >>>> > >>>> Otis > >>>> -- > >>>> Performance Monitoring * Log Analytics * Search Analytics > >>>> Solr & Elasticsearch Support * http://sematext.com/ > >>>> > >>>> > >>>> On Thu, Dec 12, 2013 at 5:37 AM, Rainer Gerhards > >>>> <[email protected]>wrote: > >>>> > >>>> Let's keep the info flowing, but one important point: doesn't that > mean > >>>> > >>>>> ASL > >>>>> is required? That's a medium problem, would require roughly 2 to 3 > >>>>> month > >>>>> of > >>>>> re-coding (best guess) features that are currently covered by GPL > code, > >>>>> only. > >>>>> > >>>>> Any thoughts? > >>>>> > >>>>> Rainer > >>>>> > >>>>> > >>>>> On Thu, Dec 12, 2013 at 11:30 AM, Rainer Gerhards > >>>>> <[email protected]>wrote: > >>>>> > >>>>> Just a tiny note due to overload ;) > >>>>> > >>>>>> > >>>>>> > >>>>>> On Thu, Dec 12, 2013 at 11:05 AM, Radu Gheorghe < > >>>>>> [email protected] > >>>>>> wrote: > >>>>>> > >>>>>> Hi, > >>>>>> > >>>>>>> > >>>>>>> I'll leave it to Otis to give more details, because he's an Apache > >>>>>>> committer, but I believe there's a misunderstanding here. > >>>>>>> > >>>>>>> An Apache project can still be backed up by a commercial company. > >>>>>>> Lots > >>>>>>> > >>>>>>> of > >>>>>> > >>>>> > >>>>> projects are, like Solr, Flume (another logging product!) and so on. > >>>>>> The > >>>>>> > >>>>>>> main advantage I see is that it's easier for the community to > >>>>>>> contribute > >>>>>>> and drive the project forward. It's a model that seems to work for > >>>>>>> open-source software, and lots of projects who got in there are > doing > >>>>>>> > >>>>>>> very > >>>>>> > >>>>> > >>>>> well - very active, growing, lots of people offering > >>>>>> > >>>>>>> consulting&support&professional services, building more complex > >>>>>>> products > >>>>>>> on > >>>>>>> top of them, packaging them in various ways, etc > >>>>>>> > >>>>>>> I think this is an idea that would help drive more contributions > and > >>>>>>> hopefully solve the 24 hours/day problem that keeps popping up > >>>>>>> lately, > >>>>>>> more > >>>>>>> and more as rsyslog gets more attention. To prove the "attention" > >>>>>>> > >>>>>>> theory, > >>>>>> > >>>>> > >>>>> let's look at some trends for rsyslog and some other products that > >>>>>> came > >>>>>> > >>>>>>> > >>>>>>> up > >>>>>> > >>>>> > >>>>> in discussions lately: > >>>>>> > >>>>>>> > >>>>>>> > >>>>>>> http://www.google.com/trends/explore#q=rsyslog%2C% > >>>>>>> > >>>>>> 20logstash%2C%20syslog-ng&cmpt=q > >>>>> > >>>>> > >>>>>> And the mailing list traffic: > >>>>>>> http://dir.gmane.org/gmane.comp.sysutils.rsyslog > >>>>>>> > >>>>>>> Maybe going to Apache would be a crazy idea in the sense that it > >>>>>>> > >>>>>>> requires > >>>>>> > >>>>> > >>>>> following that procedure. And people already have too little time. > >>>>>> > >>>>>>> > >>>>>>> Maybe a > >>>>>> > >>>>> > >>>>> less crazy idea is just to put it on GitHub or somewhere similar > where > >>>>>> > >>>>>>> > >>>>>>> it > >>>>>> > >>>>> > >>>>> would be easy to just send pull requests. > >>>>>> > >>>>>>> > >>>>>>> > >>>>>> > >>>>>> Well... done so 6 month ago ;) This is the initial blog post with > some > >>>>>> progress: > >>>>>> > >>>>>> http://blog.gerhards.net/2013/05/moving-to-github.html > >>>>>> > >>>>>> Was announced at several places. Overall feedback is "not bad, but > no > >>>>>> > >>>>>> real > >>>>> > >>>>> difference". > >>>>>> > >>>>>> Rainer > >>>>>> > >>>>>> > >>>>>> This will not only make code > >>>>>> > >>>>>>> patches easier, but I'm thinking at documentation, tests, issues, > all > >>>>>>> in > >>>>>>> one place. > >>>>>>> > >>>>>>> Not only contributions would be easier, but it would make things > >>>>>>> would > >>>>>>> > >>>>>>> be > >>>>>> > >>>>> > >>>>> more visible. Your github profile counts for many as some sort of > CV. > >>>>>> > >>>>>>> > >>>>>>> The > >>>>>> > >>>>> > >>>>> more&better you contribute, the more awesome you are (think job > >>>>>> > >>>>>>> opportunities and such). For the project, it would make it easier > to > >>>>>>> see > >>>>>>> how it evolves: contributions, issues, wiki - again, all in one > >>>>>>> place. > >>>>>>> That > >>>>>>> is familiar to many. > >>>>>>> > >>>>>>> Apache takes this idea to a higher level, as far as I understand, > so > >>>>>>> > >>>>>>> crazy > >>>>>> > >>>>> > >>>>> could be crazy-good. *Could be* - I don't know, really, I'm lacking > >>>>>> > >>>>>>> knowledge about how Adiscon and Apache works, although I have some > >>>>>>> idea > >>>>>>> about both. Otis, maybe you can say some more about the advantages > >>>>>>> (and > >>>>>>> disadvantages)? > >>>>>>> > >>>>>>> And I'm obviously curious about what the Adiscon guys have to say > >>>>>>> about > >>>>>>> this. > >>>>>>> > >>>>>>> Best regards, > >>>>>>> Radu > >>>>>>> > >>>>>>> 2013/12/11 Boylan, James <[email protected]> > >>>>>>> > >>>>>>> I agree with David. As a business who developsand supports a > >>>>>>> product, > >>>>>>> > >>>>>>>> > >>>>>>>> I > >>>>>>> > >>>>>> > >>>>> can't see any reason for Adiscon to do so unless they decide to no > >>>>>> > >>>>>>> > >>>>>>>> longer > >>>>>>> > >>>>>>> support and develop the software. > >>>>>>>> > >>>>>>>> -- James > >>>>>>>> -- Sent from my mobile -- > >>>>>>>> > >>>>>>>> ----- Reply message ----- > >>>>>>>> From: "David Lang" <[email protected]> > >>>>>>>> To: "rsyslog-users" <[email protected]> > >>>>>>>> Subject: [rsyslog] Rsyslog => Apache? > >>>>>>>> Date: Wed, Dec 11, 2013 3:09 PM > >>>>>>>> > >>>>>>>> not being an adiscon person, my first question is why they would > >>>>>>>> want > >>>>>>>> > >>>>>>>> to do > >>>>>>> > >>>>>>> that? > >>>>>>>> > >>>>>>>> David Lang > >>>>>>>> > >>>>>>>> On Wed, 11 Dec 2013, Otis Gospodnetic wrote: > >>>>>>>> > >>>>>>>> Date: Wed, 11 Dec 2013 16:05:11 -0500 > >>>>>>>> > >>>>>>>>> From: Otis Gospodnetic <[email protected]> > >>>>>>>>> Reply-To: rsyslog-users <[email protected]> > >>>>>>>>> To: rsyslog-users <[email protected]> > >>>>>>>>> Subject: [rsyslog] Rsyslog => Apache? > >>>>>>>>> > >>>>>>>>> Hi, > >>>>>>>>> > >>>>>>>>> Maybe this is a completely crazy question, but has Adiscon ever > >>>>>>>>> > >>>>>>>>> considered > >>>>>>>> > >>>>>>>> moving Rsyslog to Apache? > >>>>>>>>> > >>>>>>>>> Thanks, > >>>>>>>>> Otis > >>>>>>>>> -- > >>>>>>>>> Performance Monitoring * Log Analytics * Search Analytics > >>>>>>>>> Solr & Elasticsearch Support * http://sematext.com/ > >>>>>>>>> _______________________________________________ > >>>>>>>>> rsyslog mailing list > >>>>>>>>> http://lists.adiscon.net/mailman/listinfo/rsyslog > >>>>>>>>> http://www.rsyslog.com/professional-services/ > >>>>>>>>> What's up with rsyslog? Follow https://twitter.com/rgerhards > >>>>>>>>> NOTE WELL: This is a PUBLIC mailing list, posts are ARCHIVED by a > >>>>>>>>> > >>>>>>>>> myriad > >>>>>>>> > >>>>>>> > >>>>>>> of sites beyond our control. PLEASE UNSUBSCRIBE and DO NOT POST if > >>>>>>>> you > >>>>>>>> DON'T LIKE THAT. > >>>>>>>> > >>>>>>>> > >>>>>>>>> _______________________________________________ > >>>>>>>>> > >>>>>>>> rsyslog mailing list > >>>>>>>> http://lists.adiscon.net/mailman/listinfo/rsyslog > >>>>>>>> http://www.rsyslog.com/professional-services/ > >>>>>>>> What's up with rsyslog? Follow https://twitter.com/rgerhards > >>>>>>>> NOTE WELL: This is a PUBLIC mailing list, posts are ARCHIVED by a > >>>>>>>> > >>>>>>>> myriad > >>>>>>> > >>>>>> > >>>>> of sites beyond our control. PLEASE UNSUBSCRIBE and DO NOT POST if > you > >>>>>> > >>>>>>> DON'T LIKE THAT. > >>>>>>>> _______________________________________________ > >>>>>>>> rsyslog mailing list > >>>>>>>> http://lists.adiscon.net/mailman/listinfo/rsyslog > >>>>>>>> http://www.rsyslog.com/professional-services/ > >>>>>>>> What's up with rsyslog? Follow https://twitter.com/rgerhards > >>>>>>>> NOTE WELL: This is a PUBLIC mailing list, posts are ARCHIVED by a > >>>>>>>> > >>>>>>>> myriad > >>>>>>> > >>>>>> > >>>>> of sites beyond our control. PLEASE UNSUBSCRIBE and DO NOT POST if > you > >>>>>> > >>>>>>> DON'T LIKE THAT. > >>>>>>>> > >>>>>>>> _______________________________________________ > >>>>>>>> > >>>>>>> rsyslog mailing list > >>>>>>> http://lists.adiscon.net/mailman/listinfo/rsyslog > >>>>>>> http://www.rsyslog.com/professional-services/ > >>>>>>> What's up with rsyslog? Follow https://twitter.com/rgerhards > >>>>>>> NOTE WELL: This is a PUBLIC mailing list, posts are ARCHIVED by a > >>>>>>> myriad > >>>>>>> of sites beyond our control. PLEASE UNSUBSCRIBE and DO NOT POST if > >>>>>>> you > >>>>>>> DON'T LIKE THAT. > >>>>>>> > >>>>>>> > >>>>>>> > >>>>>> _______________________________________________ > >>>>>> > >>>>> rsyslog mailing list > >>>>> http://lists.adiscon.net/mailman/listinfo/rsyslog > >>>>> http://www.rsyslog.com/professional-services/ > >>>>> What's up with rsyslog? Follow https://twitter.com/rgerhards > >>>>> NOTE WELL: This is a PUBLIC mailing list, posts are ARCHIVED by a > >>>>> myriad > >>>>> of sites beyond our control. PLEASE UNSUBSCRIBE and DO NOT POST if > you > >>>>> DON'T LIKE THAT. > >>>>> > >>>>> _______________________________________________ > >>>>> > >>>> rsyslog mailing list > >>>> http://lists.adiscon.net/mailman/listinfo/rsyslog > >>>> http://www.rsyslog.com/professional-services/ > >>>> What's up with rsyslog? Follow https://twitter.com/rgerhards > >>>> NOTE WELL: This is a PUBLIC mailing list, posts are ARCHIVED by a > myriad > >>>> of sites beyond our control. PLEASE UNSUBSCRIBE and DO NOT POST if you > >>>> DON'T LIKE THAT. > >>>> > >>>> _______________________________________________ > >>>> > >>> rsyslog mailing list > >>> http://lists.adiscon.net/mailman/listinfo/rsyslog > >>> http://www.rsyslog.com/professional-services/ > >>> What's up with rsyslog? Follow https://twitter.com/rgerhards > >>> NOTE WELL: This is a PUBLIC mailing list, posts are ARCHIVED by a > myriad > >>> of sites beyond our control. PLEASE UNSUBSCRIBE and DO NOT POST if you > >>> DON'T LIKE THAT. > >>> > >>> _______________________________________________ > >> rsyslog mailing list > >> http://lists.adiscon.net/mailman/listinfo/rsyslog > >> http://www.rsyslog.com/professional-services/ > >> What's up with rsyslog? Follow https://twitter.com/rgerhards > >> NOTE WELL: This is a PUBLIC mailing list, posts are ARCHIVED by a myriad > >> of sites beyond our control. PLEASE UNSUBSCRIBE and DO NOT POST if you > >> DON'T LIKE THAT. > >> > >> _______________________________________________ > > rsyslog mailing list > > http://lists.adiscon.net/mailman/listinfo/rsyslog > > http://www.rsyslog.com/professional-services/ > > What's up with rsyslog? Follow https://twitter.com/rgerhards > > NOTE WELL: This is a PUBLIC mailing list, posts are ARCHIVED by a myriad > > of sites beyond our control. PLEASE UNSUBSCRIBE and DO NOT POST if you > > DON'T LIKE THAT. > > > _______________________________________________ > rsyslog mailing list > http://lists.adiscon.net/mailman/listinfo/rsyslog > http://www.rsyslog.com/professional-services/ > What's up with rsyslog? Follow https://twitter.com/rgerhards > NOTE WELL: This is a PUBLIC mailing list, posts are ARCHIVED by a myriad > of sites beyond our control. PLEASE UNSUBSCRIBE and DO NOT POST if you > DON'T LIKE THAT. > _______________________________________________ rsyslog mailing list http://lists.adiscon.net/mailman/listinfo/rsyslog http://www.rsyslog.com/professional-services/ What's up with rsyslog? Follow https://twitter.com/rgerhards NOTE WELL: This is a PUBLIC mailing list, posts are ARCHIVED by a myriad of sites beyond our control. PLEASE UNSUBSCRIBE and DO NOT POST if you DON'T LIKE THAT.
