Session Start: Mon Sep 22 15:57:00 2003 Session Ident: #kannel [15:57] * Now talking in #kannel [15:57] * Topic is 'Kannel - WAP and SMS Gateway - http://www.kannel.org' [15:57] * Set by ChanServ on Tue Sep 02 20:01:10 [15:57] Hellas [15:58] Aarno, is Andreas going to make it to this session? [15:59] I ask him [15:59] * davidtully has joined #kannel [16:02] He will soon be there [16:02] ok, great [16:02] * Malysh has joined #kannel [16:02] did you guys have a chance to review the Apache incubation process yet? [16:02] hi all [16:02] Hi Alex [16:03] Hi all.. [16:04] Paul: did you review the Apache incubation process? [16:05] * kiwi has joined #kannel [16:05] * kiwi is now known as afink [16:07] Hi Andreas [16:07] hi everyone [16:08] yes I did - it looks good, but also looks like there is a lot of things TBD... [16:09] Who will control license if we join Apache incubation process [16:10] that's a good question. Isn't this addressed in the incubation process description? [16:10] ASF will, as all software must be licensed under the Apache license [16:11] This is a condition of exit from the incubator [16:11] yep, that was my understanding too [16:12] but why ASF? [16:12] because there the legal body [16:13] as a foundation [16:13] right= [16:13] ? [16:13] so having a legal body is the only thing what we are hunting for? [16:14] no, ASF brings us also "popularity" in terms of Kannel will get "recognized" in a different way. [16:14] That's one thing; one argument for ASF is the strength of the Apache brand. Depending on what you are doing with the software, this may or may not be important to you [16:15] well, I dont know anyone who associates with the brand apache anything else than the webserver. [16:15] well, there's tomcat as well... [16:16] which no one uses *grin* [16:16] andreas: uhhh, there are a lot of projects that are *known* and because hosted by the ASF commercially respected. [16:16] like Cocoon, ie. [16:16] stipe - a clear statement on the incubator site is - This means that the codebase will need to be examined to ensure that, if and when the Foundation begins distributing it, it has clear title to do so and isn't infringing on anyone's rights. [16:16] or Axis as SOAP server [16:17] right. [16:17] andreas: we use Tomcat ;) [16:18] so if "clear title" cannot be established, its a no go. Do you know anyone at the incubator project who can give an opinion on whether "clear title" can be established on the current srcs ? [16:18] http://www.apache.org/ doesnt show Axis or Tomcat but shows PHP for exampel which is not directly a ASF project. so making Kannel a ASF project would not make it appear on that page. In other words it would not make it more popular [16:18] andreas: axis is a project in the "web services" devision of ASF. [16:19] and on php.net it is stated: PHP is a project of the Apache Software Foundation. [16:19] ;p [16:20] Has ASF wireless services section ;) [16:20] I know that PHP core developers are ASF members ;O [16:20] interesting, I never noticed that. but marketing whise I dont see much benefit. [16:20] aarno: no, that's why Kannel fits good into this place. [16:21] Kannel fits good in there, because Kannel can be considered as the "glue" between the wired internet technologies (HTTP, RADIUS, etc.) and the unwired protocolls (SMPP, WAP, MMS, etc) [16:21] IMO [16:21] we may have synegentic effects in this "open source software conglomerative approach" [16:22] Also most of us use kannel for a commercial use and we build products and services based on kannel and sell those. Kannel's current license in my eyes fits this very well. I'm not too sure about ASF's objectives in this respect. They could be counter productive. [16:22] like having the APR (apache portable runtime) to be a future base for Kannel's HTTP stack, etc. [16:22] stipe - andreas has raised a very good point - can you address it ? [16:23] andreas: Apache is sold millions of time in IBM WebSphere, Oracle Application Server, and in Covalent Commercial Apache products. [16:23] AFAIK - and I've been to several ApacheCon conventions - the ASF *wants* explicetly to have Apache a commercial grade server. [16:24] which means it has to proove it's goodness in the field. [16:24] which also means it is packaged as part of a commercial product (see above examples) [16:24] Of course, we don't want to loose the right to bundle Kannel into commercial products. [16:25] so there is no real difference between what we can do now with Kannel in commercial products and what we could with it under an Apache license ? [16:25] But that's exactly what Apache httpd developers from IBM do. They develop at apache and then bundle it to a commercial product into their IBM WebSphere application server. [16:25] paul: right! ASL is still BSD-style, which means we can bundle it to commercial products, except that we have to state where it comes from. [16:26] in terms of stating this is based on Kannel and based from a project maintained by the ASF. [16:26] I have not seen anywhere on the website of IBM when looking for WebSphere that it is based on Apache. [16:26] This has to difference to an "own" foundation IMO. [16:28] * Bill-B2 has joined #kannel [16:29] what about the contributor license agreement - I guess everyone (who contributes) has to sign this to work on an ASF incubator project ? [16:29] timeframe: setting up a KSF will take time and going for ASF would be faster was one of the arguments. [16:29] I read: [16:29] The Apache Software Foundation will eventually spread its wings to encompass more collaborative, open-source software projects. However, at the present time we have our hands full setting up the relationships among our existing projects. If you send us a proposal for a new project, we will certainly read it, but we may not have time to consider adopting it as a new ASF project until we have the resources available. [16:29] in other words we might put lots of efforts into something which might be considered, one day, maybe. [16:30] andreas: php has also been an "intependend" project and is now formed under the ASF. If don't thing we'll have too heavy formal hurdles to take here. [16:31] well we have the license issue standing out at the moment. I clearly dont see any benefit from ASF which we dont have in KSF. [16:32] andreas: IBM's HTTP server (which is the HTTP server in the IBM WebSphere Application server) is stated to be based on Apache, see http://www-3.ibm.com/software/webservers/httpservers/IHS20.html [16:33] andreas: the most benefitial is: a) developers increase, because ASF is more recognized then KSF will ever be [16:34] b) industry acceptance. ASF is a "global player" in the HTTP server market (and also spreading towards the Jave Jarkata projects) [16:34] stipe do you really think because we can put the apache logo on to kannel any developer would join this group? [16:34] * Bill-B has quit IRC (Read error: 60 (Operation timed out)) [16:34] (see Ericsson's MMSC, which is almost based fully on Apache Java libraries) [16:35] if we want to have Kannel at some time named a "commercial grade" SMS gateway (and/or MMSC), then we need to have an strong developer and industry acknoledgement background IMO [16:35] I'll make an example: [16:35] industry acceptance might be true for Apache HTTP server but for any of the other freaky apache project (well maybe not PHP which is known by itself) I dont think it increases acceptance. [16:35] I don't it will significantly boost the developer base; Kannel has been around for nearly 5 years and should be well known by those interested by now [16:36] paul: yep, but you miss one cuttial thing, that I consider as "stategic"!!! [16:36] paul: the US market is just raising in terms of SMS, WAP and MMS! [16:37] stipe: this is your own specific problem ...*grin*... [16:37] I know this well :-) [16:37] andreas: no, it's an industry problem we are facing here. [16:37] I think this would only change if we would change the name from kannel to ApacheWireless or something like that [16:38] Thats not a bad idea... [16:38] the US market will have the need for good SMS, WAP and MMS gateways, and why not have Kannel as "the" open source commercial grade product for this needs, as Apache was the HTTP server when the WWW boom started?! [16:38] -1 ;p [16:39] the ASF buys will not have you promote the term "apache" as part of a sub-project name. [16:39] Kannel may be known by now, right. But is it widely accepted?! [16:39] Seriously, if you buy into the Apache brand, then renaming makes sense - what does Kannel mean in the US anyway ? [16:39] how about ApacheWirless, a ASF project which uses Kannel sources? so a spin off? Would mean apache people use kannel as base. would even look better [16:40] nothing, but Apache (for those who know) didn't change it's name too from the beginning. [16:40] paul: right Kannel ASF or Kannel KSF makes no differenc in US market unless we can profit from the apache branding. [16:40] -------1 ;pppp [16:40] Apache didn't change, PHP didn't change, Kannel shouldn't change too IMO. [16:40] that's a tribute to the system architects that started everything. [16:41] you wouldn't like that someonce who picks up your babdy after you died would rename it, would you? [16:41] ok lets see who in this group was aware that PHP is a ASF project and would buy PHP because its from Apache where they wouldnt if its not apache? [16:41] I'd be dead, so I would'nt care ? [16:41] * stipe did know [16:42] stipe is a rare exception... [16:42] ;P [16:42] or: Stipe may be better informed ;) [16:43] we have to look at this as from the eyes from potential buyers. [16:43] Alex: any comments from your side here? Seems you're a bit silent, maybe I have to kick your ICQ butt? [16:43] ok [16:43] * stipe will hear this perspective [16:43] go on [16:44] lets say this buyer will buy stipes megabooster-sms product where it says "based on Kannel, a ASF project" [16:44] do you think this will make this product look more serious ? [16:44] andreas: I didnt know that php as asf - however I use php based on the fact its a useful language [16:44] * Bill-B2 is now known as Bill-B [16:44] yes.. and why? because Apache is associated with the Apache HTTP server, the defacto market leader. [16:45] bill, thats exactly the same with me. I used PHP because its powerful not because its ASF, I only realized that today. [16:45] When you buy an Office product (like MS Office), which one do you buy, you buy MS, because it has been proven to be most widely accepted. [16:45] stipe: does having the apache name on it mean our kannel is as rock solid? NO. [16:46] We are not talking about an addon to apache's HTTP server here, we talk about a totally independent product developed by totally independent developers. [16:46] no. I'd like to attract more developers to it. [16:46] being completely honest... till on the ML's i saw ASF and KSF mentioned, i didnt release it was a software foundation.. though I dont keep my "finger on the pulse" [16:47] * dt has joined #kannel [16:47] I think/hope apache brand will biring us more popularity and new developers (means new ideas) ... [16:47] dt: please use a nick that is close to your real-name, since we log this session. [16:47] * davidtully has quit IRC (Read error: 54 (Connection reset by peer)) [16:48] and I consider this also from a serspective where we can have synegertic effects. [16:48] i can only say , now Kannel is NOT a rock solid product (out of the box)... [16:48] Can you give us an example of possible synergetic effects ? [16:48] Ie. having Apache HTTP server as the base for large scale MMSCs, etc. [16:49] right. the synergies is important point. currently I dont see any. [16:49] * dt has left #kannel [16:49] OK, but we could do that anyway. If I was building an MMSC, it'd be Apache for the w/s and Kannel for the WAP/SMS gateway... [16:49] when you imagine that in 5 years all SMS traffic has been switched to MMS. This will mean you will need extensive HTTP traffic to be handled for this. [16:49] no sorry, no examples yet :( therefore "hope" [16:51] paul: how? We have a Kannel based MMS encoder/decoder library and Apache that does the HTTP traffic. Now you need the "glue" between that, let's call it mod_mms. This implements the Kannel MMS decoding/encoding and passed MMS to the back-end logic. [16:51] just an example. [16:51] * davidtully2 has joined #kannel [16:51] Lets not design an MMSC right now... :-) [16:51] ok [16:52] stipe: MMS doesnt strike out the need for SMS at all as every MMS requires SMS. And frankly I dont see what MMS has to do with a HTTP traffic. Kannel doesnt have a MMSC and the built in HTTP server of kannel is pretty good in terms of resource use. So the only profit would be if we build a mod_kannel for Apache and I think this is a total change of goals which we are not ready to address. [16:52] but it's an example ;) [16:52] which is totally unrelated to Kannel SMS and Kannel WAP gateway which are todays codebase. [16:52] alex: could you pass Andreas some of your gwlib/http.c bug fixes, I guess he would be amazed ;) [16:53] ;) not yet... too much work ... [16:53] ok, but don't forget ;) [16:53] frankly until now I have not see any good reason to take the jump to ASF, only hassles. [16:53] as I said: my vision is to have Kannel integrated into a more complex resource environment. [16:54] therefore i said: yet Kannel (out of the box) is not a rock solid, we ve too many bugs there... [16:54] ASF is an organization that may help provide this in an easier way. [16:54] which makes the kannel project as it is today obsolete [16:54] and get new developers involved into kannel would be great [16:54] no [16:54] andreas: why obsolete? [16:55] I strongly doubt alex that we will get any new developers because we have ASF on the page. Developers usually join project because they love coding and have a specific need to do something. They usually rarely care about ASF if they join open source projects. [16:56] andreas: to give you one example! [16:56] I send an invitation to the incubator staff at ASF, to have them join the past meeting on Friday. [16:56] and what happened? [16:56] I think this discussion exists because stipe wants to sell his stuff and needs a nice brand name on his box only ;-P [16:57] A project coordinator of the James (POP3, SMTP server) was thinking on how to integrate/connect to Kannel for SMTP-to-SMS and vice versa traffic. [16:57] so, in software markets it's like in real-asset markets. In some sense you have to attract people. [16:57] yes, most of us use PHP, because they know it since version 2.0 (at least me ;) [16:58] but, they get the "ships" that help debugging and minor code work with the "name arround it". [16:58] I meat sheeps, sorry ;( [16:58] afink: popularity of the project == new developers or at least ideas [16:59] andreas: I don't need to sell anything. My salary at Wapme is quite honorable. I'm a visionier, exactly as people that started the httpd Apache years ago to have a open source product to be the best in it's field. [16:59] no, it's not really an equation, unfortunatly it's not that easy. [17:00] it's more like a differential equation ;) [17:00] stipe: so why dont you contribute your code more to the project? your emibox you announced many times etc etc [17:00] I don't think comparsions like that between Apache httpd and Kannel are valid. Apache httpd has a critical mass that Kannel will never have. [17:00] BECAUSE I want to have that DAMNS configure/make/buld process cleanly and have these as sub-cvs modules! that's why. [17:01] paul: really??? I aiming to have most MMS traffic handled in the US and elsewhere on code that is based on Kannel. [17:01] and because a people are busy , andreas? [17:02] yep. [17:02] apache counts for about ~50% of web servers out there right - if you compare HTTP requests to SMS messages carried by live Kannel installations, you get muy point. [17:03] Apache was also used at that time by some "stupid" admins that had fun in putting a different httpd daemon on their AIX or solaris box than the CERN httpd or the NCSA httpd. And WWW was almost strictly used by unversities. [17:03] I ve there many bugfixes in our tree, but just no time to commit these in cvs :( [17:03] paul: that's right. But we would like to have Kannel in that field in SMS, WAP and MMS, right? [17:04] stipe: is ir a really question not, how would we like to intergrate Kannel into Apache ? [17:04] Commercial vendors will always exists. But (as Apache did), giving people and even large companies the opportunity to run software that is open and of commercial grade gives alternatives. That's all what it is about. [17:05] IMO, there is just a question: what is a benefit for Kannel project if we move to ASF ? [17:05] alex: I don't know exactly. BTW, sounds like people thing I'm trying to push a decission towards a certain direction (the ASF one), which is *NOT* the case. I'm just putting things into alternative discussions. [17:05] didn't I mentioned those already? [17:06] yes [17:06] * stipe scratches his head [17:06] yes, but it's just tooo many disc. about or ? [17:06] * Pulp has joined #kannel [17:06] ok, now, what benefits do we have if we move the KSF road instead of ASF? [17:06] greets to all [17:06] Lets look at it another way - what is the risk of a tree split if we don't move to ASF ? [17:06] pulp: please pick a nick that is close to your real name! [17:07] paul: nothing IMO. [17:07] paulkeogh: I like ur question more ;) [17:07] the question is: which path (we move) will bring us most velocity? [17:07] In my eyes the only thing which makes Kannel as successful in WAP/SSM as Apache in the HTTP area is the constant work of everyone to make it improve. [17:08] and the answer is: noone knows ;O [17:08] What I currently see is that everyone tries to "hide" his bugfixes or his improvement from others. [17:08] andreas: right! that's why we are here. [17:08] Personally I think this is something bad in the current state. [17:08] nop. At least not Centrium or Wapme! [17:08] * Pulp is now known as Alexei [17:08] stipe, noprob. [17:08] Alex has commited a large ammount of personal bugfixes that have been in a seperate cvs tree into the official tree. [17:09] Hi Alexei [17:09] from our side we have done quite some work on the MMS side of the WapPush and every single bit went into Kannel. In other words we run kannel umodified here. [17:09] hi all [17:09] Alex: what was the reason for this separate cvs tree? [17:09] stipe, what is your primary goal with the project? [17:09] andreas: I run Kannel also unmodified for our MMSC, it's only the mmsbox (the HTTP server and MMS logic) that is yet not included. [17:10] andreas: you didn't contribute MMSC too, and never intended to, did you ? ;P [17:10] afink: sorry, it's just too much... I try to commit _all_ things into cvs , but really ve no time for this ... [17:10] private cvs trees will always be there, some people do additional modules, sometimes totally not required by community. [17:10] right. [17:11] Personally, in by desk I have this 2-3 beasts waiting to be released, smppbox, emibox, mmsbox. [17:11] stipe: MMSC is not a kannel product, not yours, not mine. this is a different thing. Which brings us to the point if having a MMSC built into kannel would not make the kannel project more attractive for all. [17:11] maybe required , but none known it as long no need there [17:11] stipe, our mmsc is not a "part" of kannel, its just based on kannel libs. [17:11] alexei: yep. that's ok. [17:11] * afink think stipe's too busy to commit all those boxes... [17:12] * stipe thinks andreas is right here. [17:12] ;) [17:12] * stipe tries to do it anyway ;) [17:12] that's why I'm waiting for the cleaner cvs module seperation and the configure/build process where we can have "external" boxes compiled on the fly in the build process [17:13] BUT, this is actually not todays topic ---- let's get back to the road. [17:13] I think making kannel more attractive requires work on kannel, not trying to chance the license and going into some foundation which has nothing to do with kannel at all. In the case of PHP there is a direct relation to Apache somehow because PHP is a module to Apache but in our case there's no link. [17:14] ??? but we can me this link! [17:14] * Alexei wonders what "attractive kannel" means. [17:14] We dont need apache, apache doesnt need us, the only thing common is that both are open source projects. [17:14] by having Apache the HTTP server base for Kannel's MMSC. (ie). [17:14] kannel is a tool used by particular companies for specific purpose, why make it "attractive" and why is it not now ? [17:14] why does PHP need Apache? It could run as CGI? [17:15] stipe: mmsc is NOT part of kannel's code. so go and build ApacheMMSC if you wish [17:15] andreas: But it will become shortly, with the release of our mmsbox cvs module to Kannel!!!!! [17:15] another "wish" from stipe. [17:15] Consider our (Wapme developed) MMSC to be part of Kannel. [17:15] stipe, make mmsbox an apache license [17:16] nop, mmsbox is a Kannel module and hence under Kannel license. [17:16] stipe, u will have a new executable, kannel license needs no change for that, mmsbox as a new product can have a different license [17:16] yes so everyone can copy it and sell it *g* [17:16] ?? why is that? [17:17] I think we have to face one important fact when we compare to apache [17:17] Kannel license will apply to the mmsbox cvs module [17:17] The whole comparision is a light into the dark path we face, nothing more. [17:17] Apache was made as an alternative HTTP server to the Cern HTTP server and the NCSA HTTP server and turned out to be better in the end. [17:17] stipe, what is your distant future plan for kannel ? [17:17] right. [17:17] both where open source and free. [17:17] right. [17:18] Apache's main break-through was it's module API., [17:18] Kannel competes with some freaky tools which do HTTP to SMS in some more or less fancy way. [17:18] stipe, u want to ban people from using in commercial purposes with no fees? [17:18] Alexei: ? banning, I don't want to bann someone? [17:18] currently I dont see any open source project in this area. Going to MMSC means competing with COMMERCIAL guys like Nokia and Ericssons who sell their stuff for millions. [17:18] this is a totally different thing. [17:19] I don't think so. [17:19] or compare it to Linux versus Microsoft if you wish :-) [17:19] Having Kannel as SMS (and MMSC ;) in an IBM commercial product this would be a competition. [17:19] its a long road to go and the only way to make it succeed is stability & performance. [17:19] I have talked to marketing people at Nokia and Ericsson. [17:20] right. [17:20] the main things is: Nokia and Ericsson shouldn't make money by selling server software. That's not there core competence. They make money with selling phones. [17:20] lets turn it around: what would WapME, GNI, Centrium, Anam etc benefit from the fact that IBM would create a MMSC and sell it for good money? [17:21] IBM stopped their own HTTP server development because Apache was better. Why? Because selling AIX boxes is their core competence. [17:21] Yeah, but they bundle phones and MMSC together in the deal. Just ask LogicaCMG... [17:21] believe me Nokia make a lot of money by selling infrastructure to operators. [17:21] * Mat972 has joined #kannel [17:21] andreas: I wasn't talking about base station and hardware. [17:22] afink: what would be the benefit for those companies? [17:22] * stipe thinks... [17:22] a MMSC is also recognized as infrastructure as a SMSC is. its a box in a rack in some operator's rack [17:22] * davidtully2 has left #kannel [17:22] yes, but the software "could" be exchanged, if the other software is better?! [17:22] the benefit for IBM would be huge they make millions. We make nothing. [17:23] nop. [17:23] (IBM to be considered a random example) [17:23] the key is the guy who SELLS makes the money, not the guy who develops [17:23] IBM makes the deployment. We make the licensing partitions that IBM will spend us in keeping the product vital and stable. [17:23] when there is no competition money wise (Cern HTTP, NCSA HTTP versus Apache) this of course doesnt apply [17:24] IBM will spend us?? [17:24] that's not true. I can give you examples in the Apache area where the developers have been getting the money in a fair share to those who sell. [17:24] yes [17:24] and why? [17:24] because of core competence, again [17:25] a company like IBM would hire a few people to develop on this project, make it better and sell it profiting from the brand WE already established. [17:25] IBM is the system (overall) integrator. They know that they have a product called XY (based on Kannel) and it is maintained by a well organized group/foundation whatever. [17:26] nop, disagreed. IBM sees open source software not as asset! [17:26] that's the main difference! [17:26] why do they spend so much money in open source participation? [17:26] I'm still lacking the benefit for us.. [17:26] simple: they see open source as a technology, not an asset they can have in their books. [17:27] simple: we are the ones who provide the "technology", hence IBM will have an ineterst to keep us doing whatever we can do best, and this is providing a good software. [17:27] well anyway, ASF will not change anything there. [17:28] * stipe thinks that Andreas has really no "insight" on how ASF works and what kind of personal inter-relations those guys have to the industry ;p [17:29] back to the questions guys [17:29] stipe: I dont need ASF in my daily work. I try to keep my stuff as independent as possible. [17:29] * stipe would like suggest a quick vote (non-final or obligateble) who wants KSF and who wants ASF!!! [17:29] any dependency adds restrictions. Thats the base line. [17:30] andreas: you're a person that started his own business. [17:30] can someone repost history of this channel with "positive vs negative" of joining apache doundation ? (if there were such entries) [17:30] andeas: Some people who did the same with Apache (not to be named) make millions in gathering core Apache developers into their business and sell large-scale support contracts for intranet maintainance to banks and insurances. [17:30] doundation = foundation [17:32] * afink asks himself why Aarno is so silent... [17:32] andreas: if you condier the top 500 companies in the US as example, what HTTP server do they use for their intranet? and who is maintaining this as a commercial support entity? [17:32] the apache foundation? no. [17:32] yes, Aarno!!!! come on [17:32] stipe, why not have a group of support or why not sell support for kannel as it is today? [17:32] also I have not read a single word from bruno. [17:32] ok, so we have a negative vote here from Andreas. [17:33] definitively. [17:33] bruno here? [17:33] no, Bruno seems to be at lunch ;)))) [17:33] again *ggggg* [17:34] andreas: what is the threat you seem to face with ASF? [17:34] the same restrictions (to you personally) would also apply when having KSF. [17:35] not really because in KSF I know whom I'm dealing with, in ASF there are tons of external folks we have to deal with which we dont know, have never seen kannel etc. [17:35] you may be a project steering commetee member in Kannel at ASF or KSF. There will be still democracy [17:35] its like if you would merge your company with microsoft to show an extreme example... [17:35] democracy between Kannel developers or between ASF members? different thing [17:36] Also an issue might be that ASF is based on US laws... you might not sell your products to countries like Irak *g* [17:36] by the way whats the country of origin of the guys on the list? I only know of people from europe somehow. [17:38] ok, I understand Andreas here. [17:38] I agree, same with me. [17:38] But, AFAIK from my participation in the ASF (I'm the cygwin plaform maintainer for Apache HTTP server) [17:39] every project has it's own developers and project steering commitee. There will now be somehting like: Apache members tell you: ok, guys, we want you to develop in this direction. [17:39] All participating projects are still "independent" is this sence. [17:39] at least this is my understanding from the culture in the ASF. [17:40] They have some formal processes, like the voting, etc. But these are needed anyway [17:40] stipe, name a project under KSF that went this way [17:40] the steering commitee is correct but how about the license? [17:40] But they are not a company where you have a command hirarchie or something like this. [17:41] andreas: hmmm, I guess I would like to have an anchor in there (as you). That the license for kannel could only be changed with a majority of the Kannel developers and not the ASF members of the board. [17:41] alexei: ??? [17:42] stipe, do u know other project, started under BSD and moved later to apache ? [17:42] PHP I guess ;) [17:42] they are totally autonomous and only benefit from the ASF. [17:42] was PHP really an open source project before in the sense that there where many open source developers working on it? [17:42] there is no one from the "old ASF members" that drops "decission" on a PHP way [17:43] andreas, yes. [17:43] ok, let's take PHP as an example. I will outline this know. [17:43] PHP stated and developers gathered. [17:43] anyway, I would like to hear more comments/views than stipe's and mines.... [17:44] with php-3.x some of those get also invoted in Apache projects [17:44] and they realized that they could use synergy effect if they move PHP to the ASF. [17:44] mod_php is really the gule for PHP and ASF... [17:44] mod_kannel? nope. [17:44] ASF is not a "old-boys club" [17:45] Paul: any comments from you regarding the ongoing debate? [17:45] * afink thinks why the age of guys makes it more interesting [17:46] I think all the MMS coming in front of us would be the "glue" between ASF and Kannel [17:46] What about plain old WAP [17:47] including the PPG, the MMS encoder/decoder library [17:47] what you mean Aarno? [17:47] WAP, hmm, in WAP 2.0 we don't have the need for pure WSP<->HTTP convertion. I think WAP will reduce a bit, but will be still the umbrella of it all. [17:48] It would be also great (if we move to ASF) to have ASF in OMA!!!! as member and hence steer with on the wheel! [17:48] oma? [17:49] Open Mobile Appliance (the post WAP Forum) specification organization [17:49] Can we try to reach some conclusion here ? Stipe - do you want a vote now ? [17:49] Alliance, sorry. [17:49] Does ASF participate specs committees ? [17:49] KSF could do that too. [17:49] hmmm, not a final one, I guess we still need some debates to have. [17:49] But KSF couldn't afford too IMO. [17:50] if KSF can not afford, ASF can neither, WAP Forum is not interesting for any other Apache project so far [17:50] * stipe calles for a (non-final) vote, please do so by saing "KSF" or "ASF" (and beware of typos ;) [17:50] KSF [17:50] * stipe knew that ;p [17:51] ASF [17:51] * afink knew that ;-PP [17:51] KSF [17:51] (with certain restrictions) [17:51] ASF; happy with KSF too though [17:51] (unless i see more examples or list of +/- made by someone neutral) [17:52] * afink thinks its already late and has to go soon [17:52] alexei's vote is not counting fully (IMO), because not a developer. [17:53] stipe is wrong here [17:53] right, me too. [17:53] ASF, KSF if it realistic [17:53] why? [17:53] alexei developed a lot in kannel before he joined GNI. tons of his code in the tree [17:53] stipe: do u ask me? [17:53] * stipe did miss some patches or commits of Alexei? [17:53] alex: nop. [17:54] andreas: what is his cvs username? [17:54] >? [17:54] stipe : Sema [17:54] * stipe hasn't his name in mind for this. [17:54] aha [17:54] have to check ;)= [17:54] stipe, i was probably the only one working on it [17:55] but not lately, right? [17:55] after the old code was dumped when x.25 radiopads were cancelled in uk [17:55] not lately. [17:55] which means not lately since some years ;) [17:55] 1.5 y [17:55] hmmm [17:56] come on [17:56] ok, this makes currently: [17:56] users on channel are not 10% of developers [17:56] even [17:56] KSF 2 v.s ASF 3 votes (including Alexei's vote ;) [17:56] how many people have CVS access? [17:56] a dozen or so [17:56] stipe, lets reschedule voting and make it web-based with auth by username@kannel ? [17:56] more... [17:57] but some account are old and should be kicked. [17:57] I thought we want consensus [17:57] only stipe can not live with KSF... [17:57] Aarno: you didn't have a prelimiary vote. [17:57] yea, we must ve consensus here!!! [17:57] that's not true Andreas! I did more work to establish KSF then anyone else. [17:58] ok so everyone can live with KSF and some can live with ASF so we have consensus on KSF but not (yet) on ASF? [17:58] Aarno: your vote please! [17:58] which is yet not a clear head of direction. [17:58] Developers have some rigth for their code, we cannot force a decision [17:58] * stipe states again, this is not a final vote. [17:59] I think voting at this time doesnt show anything... not enough people by far. [17:59] ??? but we have to choose a way. Sitting there and picking none way is not a legal thing IMO. [17:59] I have to leave immediately unfortunately. alexei keep log of the rest [17:59] andreas: right. This is a "first round" [18:00] ok, would it help if we get some of the Apache guys boarded, so you can pick them directly on cerain issues?! [18:00] * afink has quit IRC ("rushing away") [18:00] that would be great (imo) [18:01] because some people are really affraid from the big bad ASF here ;) [18:01] just too many questions are open and apache guys can get those answered ... [18:01] yep. I'll ask Ken Coar to review the #kannel log from Friday and today and to give us some pointers. [18:02] ok, thanks a lot to everyone. Sorry for not having a final consensus! This is the burbon of democracy ;) [18:02] stipe, who is the initiator of this voting ? [18:03] me [18:03] and Alexander Malysh [18:03] mainly anyone who has been talking about KSF establishment for now almost 2 years. [18:03] we're moving on the same place. We have to move forward. [18:04] leaving as is, is not an alternative IMO. [18:04] stipe, ok main reason for any kind of changes ? to make kannel look more attractive to monsters who are willing to brand it ? [18:05] main reason: getting into the league of global players. I can't fraze that differently. [18:05] Stipe: or can we get new Kannel developers this way ? [18:05] aarno: yes, I think this is more likely that having the own KSF and have a non-changed developer base. [18:05] For instance, for WAP 2.0 [18:05] right. [18:05] ie. [18:05] stipe, why do u think there will come more developers suddenly? [18:05] wait [18:06] kannel is not php [18:06] * stipe waiting............... *zzzzzzz* [18:06] right [18:06] kannel by its nature is less demanded than php or apache [18:06] PHP is a tool! a tool to have dynamic content and logic [18:06] and ? [18:06] Kannel is also a tool! a tool to have SMSC connected and WAP and MMS traffic done. [18:07] But I do anticipate that the need for SMS and MMS gateways will drastically raise in the american states, US and latin america. [18:07] * Alexei was apparently working for netvision, first company to pay zeev suraski for php .. and licensing was never transparent in PHP case neither [18:07] I know [18:08] nothing in the open source work is purely transparent, that's an formal illusion. [18:08] zeev was willing to sell it to netvision and thats it even [18:08] after he has used it as his final project in technion (his CS degree) [18:08] but thats not the issue [18:08] right. [18:09] let's move the discussion to a later point. [18:09] i see that u are trying to put things in a way attractive for commercial use [18:09] everyone agrees? [18:09] and this is great, as it should be for benefit of all [18:09] ehm [18:09] RIGHT!!!!! [18:09] but [18:10] if there's some way u impact current users, than we should find a different way. [18:10] * stipe waits until Alexei finished this point [18:10] impact? [18:10] in terms of the license? [18:10] and as far as i see andreas concerned, businesses like gni has might be somehow impact [18:10] yes [18:10] we make products out of kannel too, u know [18:10] I'd say that there would be no impact to it. [18:10] VAST, SMSC, MMSC [18:11] Wapme has the same commercial interest as GNI! Why should I hurd the company that pays me?! [18:11] * stipe has to tention to sadoism ;) [18:11] BTW, same for Centrium. [18:12] stop, stop; Centrium doesn't sell any products based on kannel [18:12] as I said. I want benefits from the ASF incubation, not restrictions. If we agree that everyone would take the benefits and there are no cerious impacts on our businesses, then everyone would agree I guess. [18:12] stipe, what is andreas so concernet about than ? [18:13] concerned [18:13] alex: yes you do! A service is also a product. The license does not state tengible goods. [18:14] alexei: good question. He's a business owner. They think a bit differently ;) [18:14] alexi: for him it is like someone comes arround and tries to have a merger on him. [18:15] not really [18:15] but changing the licensing and joining the foundation are different things by me [18:16] alxander: so it's no difference if you bundle a box with software that is based/or includes Kannel, or if you sell a service that us using Kannel. That's the paradox thing about those licenses. Actually every WWW hosting company would have to state "our service is based on the Apache server". [18:16] if u want to protect your IP by having an apache-style license, it still does not involve some new people deciding for the project like in case of foundation joining. [18:17] they don't decide on the project. They decide on the group. [18:17] let's have some of the Apache folks that can give us some insightment. [18:17] * stipe suggests to delay debate to a later point? [18:17] fine by me [18:17] ok [18:18] see you all! [18:18] cya [18:18] thanks again! [18:18] thank u 2 [18:18] * paulkeogh has left #kannel Session Close: Mon Sep 22 18:18:27 2003