--- Log opened Sun Jun 30 12:59:02 2002
12:59 -!- Irssi: Log file ~/irc.log.Window6 opened
12:59 < mds> I'll log and make available.
12:59 <@fitzix> groovy
12:59 < FrePort> Cool.
12:59 <@fitzix> for this meeting, we should keep it just the three of us for now -- but open it up a little bit later on
13:00 < FrePort> agreed... though I wish we had Hans.
13:00 <@fitzix> I wish as well
13:00 <@fitzix> I'll send him an e-mail
13:01 < mds> yes. It would be nice to get everyone on the same page, even if the projects are fairly independent.
13:03 <@fitzix> done
13:04 < mds> k. back in one hour
13:04 < mds> ?
13:04 <@fitzix> Sure thing -- we'll catch up then
13:04 <@fitzix> thanks
13:04 < mds> k
13:04 <@fitzix> see you guys soon
13:04 < FrePort> okay. one hour it is.
13:05  * fitzix is away: shower/cleaning -- be back at 2:00 EST
13:41 -!- FrePort [~chatzilla@207-172-44-14.s14.tnt1.strt.va.dialup.rcn.com] has quit [Read error: 104 (Connection reset by peer)]
13:44 -!- FrePort [~chatzilla@207-172-44-14.s14.tnt1.strt.va.dialup.rcn.com] has joined #dotgnu-auth
13:59 < mds> bak
14:03 < FrePort> Still here, but having to explain biometric support... Does MACS support biometric devices, or could it?
14:06 < mds> yes, through easy modules
14:06 -!- FrePort [~chatzilla@207-172-44-14.s14.tnt1.strt.va.dialup.rcn.com]
14:06 -!-  ircname  : New Now Know How
14:06 -!-  channels : #dotgnu-auth #dotgnu 
14:06 -!-  server   : irc.openprojects.net [http://www.openprojects.net/]
14:06 -!-  idle     : 0 days 0 hours 2 mins 52 secs [signon: Sun Jun 30 13:42:42 2002]
14:06 -!- End of WHOIS
14:10 < FrePort> Good, that's what I told the guy (I was crossing my fingers that I was correct)
14:13 < mds> heh. Good guess. =)
14:13 < FrePort> Any word from Hans?
14:13 < mds> I haven't heard any...
14:17 < FrePort> Fudge.
14:32  * fitzix is back (gone 01:26:53)
14:32 <@fitzix> sorry about that
14:32 <@fitzix> took a little bit longer than I thought
14:33 <@fitzix> no word from Hans yet on my e-mail
14:33 <@fitzix> everyone here?
14:36 < FrePort> Yes, I'm here.
14:37 <@fitzix> cool - mario?
14:37 < mds> aye
14:37 <@fitzix> good
14:37 <@fitzix> OK - we're taking a look at the current state of the auth projects
14:38 <@fitzix> so far, we've got three leading auth projects -- Freport, IDSEC, and MASC (alphabetical order)
14:38 < mds> It's actually MACS.
14:38 <@fitzix> each one has it's benefits, and also it's detractors
14:38 <@fitzix> sorry - typo
14:38 < mds> np =)
14:39 <@fitzix> just so that we're all on the same page, do each of you want to give a synopsis of the basic structure and methodology of your projects?
14:39 < mds> sure. FrePort: go ahead?
14:39 <@fitzix> also, a workup of how many developers you have working with you (for resource assessment)
14:40 < FrePort> The FrePort project is best summed up by the documents at http://users.ids.net/~nightspd/freport/
14:41 < FrePort> The basic architecture design is complete. First code release is being delayed by the dependent libraries (OpenCL and CommonC++). Changes in the APIs to these libraries have forced my "team" (all three of us) to rewrite on a monthly basis.
14:42 < FrePort> We want something with basic local Auth/Acquire functionality before release.
14:43 < FrePort> The current developers are both students at the certificate level.
14:43 < FrePort> Your turn Mario.
14:44 < mds> Unlike FrePort, macs wasn't designed from the beginning with DotGNU in mind.
14:45 < mds> It's meant to be an easily extensible, easily embeddable solution.
14:45 < mds> At the moment, there are 2 developers working full time on it, including myself.
14:46 < mds> There is also a business side to the project.
14:46 < mds> We are organizing a corporation to sell implementation and maintenance of macs.
14:46 < mds> Though the software is GPL'd in its entirety.
14:47 < mds> We're chasing beta, and have Sept. 1 as the target date.
14:48 < mds> We've got one person to do the documentation, but the user tools aren't out yet.
14:48 < mds> That's about it...
14:49 <@fitzix> OK - just looking through the FrePort documents at the moment
14:49 <@fitzix> Mario - do you have a summary link -- for posterity in the log
14:49 < mds> Sure. The project is hosted at SourceForge: http://macs.sf.net/
14:50 < FrePort> I should add that FrePort is currently LGPL'd though I'm considering GPL with linking exception.
14:51 <@fitzix> OK - that's good on both counts
14:51 < FrePort> fitz can explain the politics basically. I have to fix the Anaconda before she strangles me.
14:52 <@fitzix> Now - the Steering Committee has, at my urging and at the urging of others, been looking at pulling in the projects to become more related to DotGNU
14:52 <@fitzix> is this idea agreeable to both of you?
14:52 <@fitzix> ouch - occupational hazard with potentially deadly snakes: death :(
14:53 <@fitzix> we're looking to make projects "officially" DotGNU projects and for those projects to get full-on endorsements
14:54 <@fitzix> There is a problem, though
14:55 <@fitzix> the "horserace" scenario that we had implemented (and I originally opposed) calls for the endorsement of any project that meets the criteria
14:55 < mds> Is that bad?
14:55 <@fitzix> which means that we're left with three auth projects
14:55 <@fitzix> which complicates things
14:55 <@fitzix> and further, one of the scenarios for the end of the horserace calls for endorsing only one project in the event that endorsing more than one overcomplicates things
14:56 <@fitzix> don't look at me -- I opposed the scenario :)
14:56 < FrePort> mds: we appear un-united and people take this as a sign of weakness: uncertainty=doubt=fear.
14:56 <@fitzix> anyway
14:56 <@fitzix> not only that
14:56 <@fitzix> but also there are practical problems
14:56 <@fitzix> we'll almost certainly have to endorse one over the others, even if all are endorsed -- just so that we can put together a package for users
14:57 <@fitzix> it creates complication for the users, which could be a serious detractor to adoption
14:57 < mds> sure, with the others as specialty alternatives, say.
14:58 <@fitzix> it also creates overcomplication in SEE as the ability for SEE to authenticate use of code has to be abstracted behind another API set -- so that we can deal with different auth plugins
14:58 <@fitzix> which may not be a bad thing in and of itself, but it introduces issues
14:58 <@fitzix> also, there is the general issue of duplication of effort.
14:58 < FrePort> LOTS of issues.
14:58 < mds> right.
14:58 < mds> doh! brb
14:59 <@fitzix> Now that we're coming to the point where we can endorse projects -- we've hit the problem scenario that I originally described to the coreteam:
14:59 <@fitzix> what if the horserace doesn't produce a clear winner? (which was bound to happen, if you ask me, that it wouldn't)
14:59 <@fitzix> what do we do?
15:00 <@fitzix> well, the first stage, IMO, is to find areas of overlap in the projects and merge them into combined projects
15:00 <@fitzix>  but, the best way to do this is to try to merge the projects themselves (since there is considerable overlap)
15:00 <@fitzix> that's my take on this
15:00 <@fitzix> considering your two designs, is there any way to merge your work?
15:01 <@fitzix> is there any way that the two designs could be made complimentary in the same package?
15:01 <@fitzix> I'll brb - gonna get some water :)
15:02 < FrePort> mds: now is that a question or what?
15:03 <@fitzix> back
15:05 <@fitzix> now - this is a brainstorming session -- it does not have the wait of an SC decision by any means
15:05 <@fitzix> so, no pressure :)
15:05 < FrePort> mds: any ideas?
15:06 <@fitzix> he's still away, I think
15:07 < FrePort> We'd best wait on him. He's furthest along on a lot of issues.
15:08 < mds> bak
15:08  * FrePort waits for Mario to read what he missed.
15:08 <@fitzix> My biggest concern is that I don't want any of the auth projects to get "left out" in any way
15:08 < mds> ok.
15:08 <@fitzix> you're all doing wonderful work and have great designs
15:09 < mds> I have 2 thoughts.
15:09 <@fitzix> I don't want to see any of that get disregarded -- not to mention it makes the SC and the project look bad to throw away the work of our members
15:09 <@fitzix> ok
15:10 < mds> first, the auth system used *by* the project doesn't preclude that other auth systems be used by applications written *for* the DotGNU.
15:11 < mds> so, for example, an DotGNU ERP system might use macs to auth its people, while DotGNU itself might use FrePort to auth the ERP system's code itself.
15:11 < mds> make sense?
15:11 <@fitzix> yes, that makes sense
15:12 < FrePort> Sort of... how about a few examples?
15:12 < mds> ok. macs has the ability to, eg, auth people using their yahoo login/passwd
15:14 < mds> Now, just because DotGNU protects its sandbox using FrePort internally, there's no reason why an application can't base its user authentication on macs.
15:14  * FrePort nods.
15:14 <@fitzix> yep - that is true
15:15 < mds> what we're talking about here is DotGNU's internal sandbox auth, though, which leads to my second thought:
15:15 <@fitzix> so, we can negate the practical problem of abstracting the SEE auth interface
15:15 < mds> right. the SEE auth interface is the sandbox-type protection?
15:15 <@fitzix> yes
15:16 <@fitzix> to authenticate the use of code by storing hashes and certs
15:16 < mds> right. that's what I thought.
15:16 <@fitzix> well, that's just one functionality
15:16 <@fitzix> that's something that a generic secure profiling system is good for
15:17 < mds> macs is flexible enough to do that, I think. But like I said, it wasn't designed specifically for that. (macs *is* a generic secure profiling system. =)
15:17 <@fitzix> precisely -- both projects will do well at it :)
15:17 <@fitzix> I have no doubt about either of your designs
15:18 < FrePort> One moment.
15:18 < mds> Now from what little I know about FrePort, it shouldn't be difficult to plug macs in to DotGNU as a FrePort plugin
15:18 < FrePort> interesting...
15:19 < mds> So that FrePort's plugin framework can serve as the general SEE auth abstraction for DotGNU.
15:19 <@fitzix> and you can get access to MACS auth facilities via FrePort's API
15:19 < mds> I can't tell right now whether that would be a good idea, but it wouldn't cost FrePort anything, since it's going to be implemented that way anyway
15:20 < mds> Yes. macs was designed, first and foremost, to be easily embeddable.
15:20 <@fitzix> and, it appears, could have other items easily embedded into it
15:20 < FrePort> mds: is there a diagram of the Macs network infrastructure. (similar to this one for IDSec? http://idsec.sourceforge.net/phpdemo.gif)
15:22 < mds> not right now. (at least, not one that's usefully up to date.)
15:22 < mds> it's something we're working on, along with a demo &etc.
15:22 <@fitzix> Do you guys think that there would be a benefit to merging, at the very least, parts of your projects?
15:23 <@fitzix> there is no right/wrong/expected answer to that :)
15:23 < FrePort> There's always benefits, but there is one majour barrier. We all wrote in different languages.
15:23 < mds> for macs, I think it's too early to tell.
15:24 <@fitzix> the languages issue can be "gotten around" -- but some work would have to be thrown out
15:24 <@fitzix> the other major barrier is the "too many cooks in the kitchen" area
15:24 < mds> there's surprisingly little code to macs, I don't think language would be a major issue.
15:24 < FrePort> very true. Counting IDSec there's 8 primary developers in 3 projects.
15:24 < mds> the cooks question is more difficult.
15:24 <@fitzix> I don't see much of a problem there between MACS and FrePort since your base design is somewhat similar in concept
15:25 <@fitzix> yep
15:25 < mds> the approaches are fairly different, though.
15:25 < FrePort> Hey, I've an idea.
15:25 < FrePort> May I have the floor?
15:25 <@fitzix> absolutely
15:25 < mds> sure, go.
15:26 < FrePort> The MACS cncept basically divides into an auth system to bring in data stored locally/remotely?
15:27 < FrePort> In FrePort this is handled by three plugin systems - configuration, local-databank, remote-databank.
15:28 < mds> umm. I don't quite understand. MACS ties to data stores using so-called "method clients." Remote or local is irrelevant except within the context of a particular method.
15:29 < FrePort> Exactly.
15:29 <@fitzix> Think of the "remote databank" as a cluster of remote "method clients"
15:29 < FrePort> That's the problem i was about to point out, but not nearly so eloquently.
15:30 <@fitzix> I definitely see merging points there
15:30 < mds> OIC. For macs, all methods are "remote." They are opaque and identical to the central system.
15:30 < FrePort> Whereas Freport uses the same API for local/remote databanks, but requires that the remote really be remote.
15:31 <@fitzix> if it's there at all
15:31 < FrePort> yes, if it's tere at all.
15:31 < mds> I see. MACS doesn't make that distinction.
15:31  * fitzix is thinking
15:32 <@fitzix> this is something I hadn't thought about originally when I'd thought of tokens... multiple remote databank storage locations...
15:32  * FrePort is thinking while grazing.
15:32 < mds> It's up to each individual "method client" to do The Right Thing.
15:32 <@fitzix> let me grok this for a minute or two
15:32  * mds is pretending to try to think and failing. ;-)
15:32 < FrePort> mds: would you define the duties of a method client for me?
15:34 < mds> They interpret a data store and answer questions (such as "Is this login/passwd combo valid?") based on it.
15:37 < FrePort> Do you mean the login to MACS or a login that MACS extracts from the datastore for a Service Provider (eg: Yahoo)
15:38 < mds> The datastore only. Method clients don't know about macs internals, generally.
15:40 < mds> The question is asked my macs' core, and action is taken based on the answer.
15:41 < mds> But my take is that only macs' "consumer" clients need to be embedded.
15:41 < FrePort> Hmmmm... I never thought I would ever say "I really need a diagram of this", but I really need a diagram of this. :-)
15:42 < mds> yeah. It's a lot trickier than I'm painting it, especially in the details, because macs uses macs as its auth system. So lots of self-referencing, headache-causing, seeming loops.
15:42 kalahari [~lgm@dsl092-008-231.sfo1.dsl.speakeasy.net] requested CTCP PING from mds: 586565195
15:43 < mds> note: kalahari is Blake Mitchel, my macs partner in crime. =)
15:43 < mds> can he join in and listen?
15:43 <@fitzix> sure thing
15:43 -!- kalahari [~lgm@dsl092-008-231.sfo1.dsl.speakeasy.net] has joined #dotgnu-auth
15:43 < FrePort> okay by me.
15:44 <@fitzix> John, are any of your people here?
15:44 <@fitzix> hello kalahari - how are ya?
15:44 < kalahari> god
15:44 < kalahari> er good even ;)
15:45 < mds> kalahari: do we have a diagram of macs general layout that doesn't have pterodactyl claw-marks on it?
15:45 < kalahari> no
15:45 < kalahari> that would be a good thing to have...
15:46  * FrePort quite agrees.
15:46 < mds> Ok. We're coding like crazy in a mad rush towards beta. General-purpose documentation has suffered.
15:47 < kalahari> you want me to whip something up right quick? what should it intail?
15:47 <@fitzix> cool :)
15:48 < mds> yeah. I think a graph of all the connection points, and a description of each connection would be enough for now.
15:48 < FrePort> http://idsec.sourceforge.net/phpdemo.gif something along this depth or this:  http://users.ids.net/~nightspd/freport/FrePort_Dataflow_Explanation_Retrieve.html
15:48 < mds> You can base it on the protocol spec, I bet.
15:48 < kalahari> ok, brb
15:49 < mds> Yes, it's nice to have a couple of full-timers. =)
15:50 < FrePort> mds: very true. I have to be satisifed with the ones I can scrape up.
15:51 < mds> FrePort: not a terrible position -- you have plenty to be satisfied with!
15:52 < mds> So.. the basic idea is that a macs client is embedded into a system. (Say, DotGNU either directly or via FrePort's plugin framework.)
15:53 < mds> These clients are configured to ask macs for the right auth/user information at the right time.
15:54 < mds> (By "macs" I mean the core macs services.)
15:55 < mds> The macs core may need to query one or more methods to answer the client's questions.
15:55 < mds> The client takes appropriate action.
15:56 < mds> Multiple macs'd systems may connect to the same macs core.
15:57 < mds> So that the macs core behaves as a sort of clearinghouse for authentication, authorization, and profile requests.
15:59 < FrePort> Well, pending that diagram. Let me see if I have this straight?
15:59 < FrePort> The SEE acts as the enviroment for establishing trust.
15:59 < FrePort> The API to the auth mechanism is through the SEE to FrePort and then...
16:00 < FrePort> through the FrePort subplugs to MACS or IDSec or native, or any other source that will or can be written to the Freport API?
16:01 < mds> Yes, if we decide that the FrePort API could be the general auth abstraction layer for SEE.
16:02 < mds> There's no reason why the macs clients couldn't be embedded in the SEE directly, for example, if Andromeda is what I think it is...
16:02 < mds> But a general abstraction for SEE's auth would alleviate some of fitzix's questions.
16:03 < mds> Do you think the FrePort API would make a good generic abstraction for SEE's auth stuff?
16:04 < FrePort> FrePort would probably make a good generalized abstraction as it is extremely flexible... the real problem is shoehorning the Macs structure into the Freport API.
16:05 < mds> That's exactly what macs is good at. =) I don't think it'll be a big problem.
16:05 < FrePort> FrePort does demand that specific functionality be handle by specific classes of plugin.
16:06 < mds> I don't think that's a problem. The macs client functionality can come in several smaller packages.
16:06 < mds> For example, the Apache plugin works this way.
16:07 < FrePort> Explain the purpose of the Apache plugin?
16:07 < mds> OTOH, some API's don't provide (much less enforce) more than one auth hook, so the macs client is more monolithic in that case.
16:07 < mds> To protect URI's via macs. 
16:08 <@fitzix> mds: I will be happy with whatever you guys decide to do -- the answer to my questions is simply that we all make an effort to work together, share ideas and share code. :)
16:08 <@fitzix> beyond that, we'll make a way out of the political problem
16:09 <@fitzix> and I think that what we have here today is a VERY good beginning
16:09 <@fitzix> we should institutionalize this
16:09 < mds> I think FrePort and macs can work together very well.
16:10 < mds> However, I'd like to concentrate on reaching beta (hence my disappearance for the last few months.)
16:10 < FrePort> I agree.
16:10 <@fitzix> me too -- and we should try to minimize work
16:11 <@fitzix> mds: sure thing -- but let's try to keep this in mind
16:11  * mds listens
16:11 < FrePort> mds: what are your dependent libs?
16:11 <@fitzix> and let's actually form the DotGNU Network Services Working Group
16:11 <@fitzix> it got nixed last time due to problems with getting resources
16:11 <@fitzix> I now have those
16:11 <@fitzix> so, this is a good beginning
16:12 <@fitzix> let's make it stick
16:12 < mds> FrePort: I've been using gdome, but it's huge. I'm looking for an alternative, or thinking of implementing my own special-purpose equivalent.
16:13 < FrePort> So, what are these resources?
16:14 <@fitzix> FrePort: originally it was just a savannah project space -- which I had to jump through hurtles to get... the FSF originally just wanted to put actual projects on it... it was more a lack of response that blocked me than actual blocking from the FSF though
16:14 <@fitzix> actually, this would more accurately be the auth working group
16:14 <@fitzix> dg-authwg
16:15 < FrePort> "dog auth wag"?
16:15 < mds> lol
16:15 <@fitzix> by the time I got it I was fighting to save my relationship with my gf so I was preoccupied
16:15 < FrePort> That's almost as bad as DUKPD.
16:15 <@fitzix> lol
16:15 <@fitzix> wag twice for yes, once for no
16:16  * mds likes simple APIs like that one. =)
16:16 <@fitzix> I watched the dg-authwg and accidentally stepped in DUKPD
16:16 <@fitzix> :)
16:16 < FrePort> ouch. I feel sorry for the duck's pud.
16:17 <@fitzix> the format is because of a limitation on the size of a name of a savannah project
16:17 <@fitzix> dg == dotgnu
16:17 <@fitzix> - == extendor
16:17 <@fitzix> and the rest should be self-explanatory
16:17 <@fitzix> FrePort: lol!
16:18 < mds> There isn't a problem with the fact that macs is an SF project, is there? I started it before savannah existed, AFAIK.
16:18 < FrePort> okay, so there's a mailing list.
16:19 <@fitzix> no - that's fine
16:19 <@fitzix> although we'd like the CVS tree to be moved if it's to be an official project
16:19 <@fitzix> you
16:19 <@fitzix> d also have to agree to be a GNU maintainer
16:20 <@fitzix> FrePort: not yet -- but it's a formality -- it *WILL* be there
16:20 <@fitzix> or, I'll e-mail bkuhn and RMS and complain :)
16:20 <@fitzix> that's actually the quickest way to get something done
16:20 <@fitzix> :)
16:21 <@fitzix> but - I have no doubts... Loic is a good guy
16:21 <@fitzix> GNU maintainer guide:
16:21 <@fitzix> http://www.gnu.org/prep/maintain_toc.html
16:22 < FrePort> May I jump the gun a bit? Talk marketting of Auth for a moment?
16:22 < mds> I'm not sure macs *should* be a GNU project just yet. As I mentioned, I'm trying to make some money back from this thing. =)
16:23 < mds> The maintainer-guide is the sort of question I'd like to hold off answering until I can spare enough time from 16-hour hacking sprees to consider carefully.
16:23 < FrePort> which segues neatly into marketting.
16:23 <@fitzix> mds: Heh - being a GNU project won't stop you... but, I can't make you be an official project :)
16:24 <@fitzix> mds: Absolutely -- no pressure... I was just making it known
16:24 <@fitzix> FrePort: absolutely - go on
16:24 < mds> fitzix: clear. Thanks.
16:25 < mds> FrePort: shoot.
16:25 < FrePort> Aside: FrePort can't be until 1) I'm out of contract 2) all the current people (angel and prowler) agree
16:25 < FrePort> We all hold joint copyright curently.
16:25 < FrePort> Anyhow, marketting...
16:26 < kalahari> ok, it's crude, but try this: http://barkingspoon.com/macs.png
16:26 < FrePort> We eventually need to identify at least five projects which require an auth mechanism, which are Free and will allow us to build a DotGNU auth mechanism into their codebase. Currently there's only one...
16:27 < mds> Which is the one?
16:27 < FrePort> kala: can you scale by 50% and repost? (it won't quite make it on a US letter.
16:28 < FrePort> PHP Groupware.
16:28 < mds> FrePort: kalahari just took off...
16:28 < mds> I'll scale it.
16:28 < kalahari> the dia file is at http://barkingspoon.com/macs.dia
16:28 < kalahari> I gotta run
16:29 < kalahari> back in a few
16:29 <@fitzix> later kalahari - nice meeting you
16:29 < kalahari> 90 mins or so that is
16:29 < kalahari> likewise
16:29 <@fitzix> thanks
16:29 -!- kalahari [~lgm@dsl092-008-231.sfo1.dsl.speakeasy.net] has left #dotgnu-auth ["Client Exiting"]
16:32 -!- kalahari [~lgm@dsl092-008-231.sfo1.dsl.speakeasy.net] has joined #dotgnu-auth
16:32 < kalahari> never mind, my sentence was commuted
16:32 < FrePort> I got the dia and scaled it from here.
16:32 <@fitzix> lol
16:32  * FrePort chuckles.
16:32 < FrePort> So, anyhow. I was thinking of a few possible targets.
16:33 < FrePort> PHP groupware targets enterprise auth users.
16:33 < mds> How about a popular webmail package?
16:34 < FrePort> Savannah could be dotGNUsed (inventing English as I go along) to target developers.
16:34 < FrePort> mds: true. More enterprise users. What package though?
16:35 < mds> IMP is popular.
16:35 < kalahari> squirrel mail is huge and supports plugins
16:35 <@fitzix> We were just going over IMP on #dotgnu
16:35 <@fitzix> IMP and Horde may be possible additions
16:36 < mds> fitzix: as in, rewrite them in C#?
16:36 < FrePort> And I'd like to target one that' used by those who aren't computer savvy - teach a "child" the value of privacy and freedom and they will accept no substitutes.
16:37 < mds> Hmm. Maybe a perusal of Freshmeat is in order?
16:38 < FrePort> mds: 60 results on webmail.
16:39 < mds> But most of them aren't terribly popular... 
16:39 < kalahari> yeah, the webmail scene is brutal these days
16:39 < FrePort> filtering by GPL...
16:39 <@fitzix> mds: no - just in their normal form
16:40 <@fitzix> mds: it doesn't have to be C# to be a webservice :)
16:40 < kalahari> much code being flung all over the place, massive code replication
16:41 < mds> fitzix: but to run "in" DotGNU, C# is all we have for now, no?
16:41 < FrePort> 18 projects in GPL or LGPL.
16:41 <@fitzix> mds: No - not exactly.  We can use any Free Software that we wish to to construct webservices.
16:42 <@fitzix> or create it
16:42 <@fitzix> that's why PHPGW can be a DotGNU project
16:42 <@fitzix> much of .Net (and other webservice tech) already exists in GNU/Linux
16:42 <@fitzix> no reason to fix/ignore that which isn't broken
16:43 < FrePort> most are written in PHP or PERL (not surprising), only one is C++ and only one in Java. None in C#.
16:44 < FrePort> Anyhow, that's the admitedly cursory results from Freshmeat.
16:44 < mds> fitzix: Hmm. I'm confused. Can we talk about it later?
16:44 < FrePort> Good news is most of them are stable.
16:45 < FrePort> So, getting back to the question?
16:45 < mds> FrePort: A more thorough browsing could give us more ideas besides webmail packages.
16:45 < mds> Eg, you mentioned educational?
16:45 < FrePort> Yes.
16:46 < FrePort> Or at least, used by teens. such as Brad Fitz's Livejournal...
16:47 < mds> Perfect. "Protect your 'blog from your parents with DotGNU!"
16:47 <@fitzix> mds: Sure thing -- I have to go soon (GF)... but if you'd like to send me an e-mail
16:47 < mds> fitzix: ok. Thanks.
16:47 <@fitzix> or we can arrange a time to get together
16:48 < mds> fitzix: mail is fine. I'm sure you'll set me straight pretty quickly. =)
16:48 <@fitzix> mds: Heh - anything I can do to clarify things -- never hesitate to ask :)
16:49 <@fitzix> It's amazing how many perceptions there are out there about this project... I never cease being amazed
16:49 < mds> fitzix: I'm sure it's just that I've got a lot of catching up to do...
16:49 < FrePort> Can we set a meeting time for next week?
16:50 < FrePort> Something with some semi-permanence.
16:50 <@fitzix> FrePort: sure thing -- what time is good for you guys?
16:50 < mds> FrePort: sure. where are you? I'm in EST. (GMT-5)
16:50 <@fitzix> EST as well
16:50 <@fitzix> that makes things easy
16:51 < FrePort> Also in EST here.
16:51 <@fitzix> weekend or weekday?
16:51 < FrePort> Virginia US of A.
16:51 <@fitzix> CT, USA
16:51 < mds> Miami, FL, Cuba. ;-)
16:51 <@fitzix> lol
16:51 < FrePort> Weekend definitely. I have to be able to make a good claim that any discussions are not on NASA's books.
16:52 <@fitzix> works for me -- next Sunday? 2:00 PM EST?
16:52 < mds> How about Sunday at 2? It was good for everybody this week...
16:52 < FrePort> And Hans (unless he moved) is at +3 GMT
16:52 <@fitzix> lol -- great minds...
16:52 < mds> great minds?!? where? ;-)
16:53 <@fitzix> for him - 2:00 PM EST is 10:00 PM
16:53 <@fitzix> mds: under a rock somewhere - I don't know :)
16:53 < mds> right. So a little earlier, then?
16:53 < FrePort> 1300 is good for me.
16:54 <@fitzix> 1300 works for me as well, how about you Mario?
16:54 < FrePort> noon isn't for obvious reasons (I'll be tangled up in snake-feeding frenzy)
16:54 < mds> Ok. 1pm (for us non-NASA types) is good.
16:55 <@fitzix> lol - don't want angry snakes
16:55 < FrePort> Is it good for kala?
16:55 < kalahari> that's 4 for me, i should be awake by then
16:55 <@fitzix> 4 AM or PM?
16:55 < kalahari> pm
16:55 < mds> yeah. he's a bum. He can do anytime. ;-)
16:55 <@fitzix> where are you?
16:55 < kalahari> san francisco
16:55 <@fitzix> very cool
16:55 < FrePort> I'll see if the prowler will be back and invite him to lurk.
16:56 < kalahari> yes, it's a balmy 82 today
16:56 <@fitzix> should I submit a proposal for the dg-authwg savannah project? 
16:56 < FrePort> Sunday then... 1300... locked, oloaded and raring to go.
16:56 < FrePort> I was about to ask if you would?
16:57 <@fitzix> sure thing
16:57 <@fitzix> I'll do it now
16:58 < FrePort> Post a news item once the savannah space is set-up. Meeting time and location...
16:59 < FrePort> So, next weeks' agenda? (loosely defined)
17:00 < FrePort> 1) Get Hans on-board with integration? (mds, fitzix)
17:00 < mds> IDSec, if possible.
17:01 < FrePort> I plan to stay out of one, reduced emnity.
17:01 < FrePort> 2) Possible market applications, what and whom would we target (all)
17:02 < FrePort> anythingelse?
17:03 < mds> I've been thinking a lot about ROI lately. But that might be best left for a later stage?
17:03 < FrePort> return on investment?
17:04 < mds> Yes. A fairly important part of any convincing argument, I think.
17:05 < FrePort> I agree. Both Barry and I have put some thought into it as well.
17:05 <@fitzix> it's important to fund all free software development
17:05 < mds> Excellent. So maybe a little of that next week?
17:05 <@fitzix> the community can't be expected to do everything for free
17:05 <@fitzix> absolutely
17:06 <@fitzix> I'm starting my own FS company so it's a very open idea for me
17:06 < FrePort> 3) ROI. (all, presentation by ????)
17:06 <@fitzix> dg-authwg registration has been submitted
17:06 <@fitzix> I'm awaiting confirmation of the project space
17:06 <@fitzix> it might take some days
17:06 < mds> cool. Let the dogwagging begin!
17:06 < FrePort> mds: you want to be the ops for that segment?
17:07 <@fitzix> lol
17:07 < mds> FrePort: segment?
17:07 <@fitzix> the code area is to be a scratch space
17:07 <@fitzix> and we'll also have a website space for presentation
17:07 <@fitzix> this way, we can have an FrePort, IDSEC, and MACS section
17:08 <@fitzix> and you guys can put diagrams, docs, and example code in there
17:08 <@fitzix> I wouldn't put actual project stuff there -- keep your own project repositories for that
17:08 < FrePort> I was thinking that we'll need to mirror our webspaces, which is good for me. That site of mine is soon to depart (changing providers after 12 years)
17:08 <@fitzix> that can be done here as well
17:09 < mds> That's a great idea.
17:09 <@fitzix> if they shit on me for it -- I'll argue it's for referrence purposes :)
17:10 < mds> fitzix: are they really that hard up for drive space?
17:10 <@fitzix> today -- we've taken our first step into a larger world
17:11 < FrePort> Definitely...
17:11 <@fitzix> mds: I don't think so -- they're not as strict as they used to be with savannah accounts
17:11 <@fitzix> but, I've heard some stories in the past...
17:11 <@fitzix> believe me, it's nothing I can't get past
17:11 < FrePort> and Mario is taking the presentation for #3; so, that squares that away.
17:11 <@fitzix> as long as only Free Software code is available
17:12 < mds> Right, ok. 
17:12 <@fitzix> and there can't be any proprietary dependancies for code on the site
17:12 <@fitzix> that's about it
17:12 <@fitzix> oh yeah - there can't be any advertisement for any prop. software
17:13 < mds> no problem.
17:13 < mds> Cool. Time to mold life out of the primordial Auth soup.
17:14 <@fitzix> if we divide the labor well, we'll get there faster with 8 people than we will with 3 here, 3 there... etc
17:14 <@fitzix> mds: precisely!
17:14 <@fitzix> mds: even if we have 3 seperate projects working together, we're a lot closer than otherwise :)
17:15 < mds> In the spirit of full disclosure, I want to emphasize that macs has an existence outside of DotGNU. I'd love to help DotGNU, and make macs better at the same time.
17:15 < mds> But it would take some convincing to get me to make macs depend on DotGNU.
17:15 <@fitzix> that's wonderful - and I want to help MACS in return
17:16 < FrePort> In the spirit of full disclosure, I have to admit that Freport was designed for DotGNU, in that I'd not have even thought of it had Norbert and Barry not repeatedly hit me over the head to document it.
17:16 < mds> Though, I'm confused about what it means to DotGNU-ify a project, and that might not even make any sense.
17:16 <@fitzix> well, maybe MACS doesn't have to "depend" on DotGNU
17:16 <@fitzix> FrePort: heh :)
17:16 < mds> fitzix: exactly.
17:17 <@fitzix> mds: That makes perfect sense to me
17:17 <@fitzix> mds: DotGNU-ify a project means two things:
17:17 < FrePort> also makes sense to me (now that I can print this *BIG* diagram.
17:17 <@fitzix> 1) the SC endorses the project as being a component of the DotGNU system
17:17 <@fitzix> 2) The project becomes a GNU project -- either under the auspices of the DotGNU project or the GNU project itself
17:18 < kalahari> FrePort: yeah, what can I say, I'm a coder not an artist ;)
17:18 <@fitzix> Now - you don't have to be dependant on any DotGNU tech to be a DotGNU project
17:18  * fitzix says to himself "we aren't GNOME" "we aren't GNOME"
17:18 <@fitzix> :)
17:19 <@fitzix> if I repeat it, maybe it'll stick :)
17:19 < mds> fitzix: but I have to interoperate... oic.
17:19 < mds> I get it. "We aren't gnome."
17:19 <@fitzix> mds: lol - yep ... no crazy dependancies
17:19 <@fitzix> but, it has to be webservice related
17:19 < kalahari> "these are not the programs you are looking for......move along"
17:19 <@fitzix> and MACS certainly qualifies
17:19 <@fitzix> kalahari: LOL!
17:20 <@fitzix> There are some things that go along with being a GNU maintainer and a GNU project
17:20 <@fitzix> I invite you to take your time in deciding
17:21 < mds> Thanks. I'll take you up on that. =)
17:22 <@fitzix> but, do work with the other projects in the meantime -- we can all make our projects better
17:22 <@fitzix> and in doing so -- maybe we'll be able to merge as the designs come closer together
17:22 <@fitzix> and maybe we'll all decide that that's the best thing
17:22 <@fitzix> if not, then so be it :)
17:22 <@fitzix> we'll find a way to make it all work
17:23 < mds> absolutely. I just want to make sure I don't promise prematurely. I have every intention of working with FrePort and IDSec and DotGNU. =)
17:23 < mds> So let's ingest today's talks. I'll make the logs available, post a url to the auth list.