jamest (jamest@fh-dialup-201091.flinthills.com) left irc: "[x]chat"
nickr (~panphage@e-172-IP28.empnet.net) joined #gnuenterprise.
psu (psu@manorcon.demon.co.uk) joined #gnuenterprise.
psu (psu@manorcon.demon.co.uk) left irc: Client Quit
psu (peter@manorcon.demon.co.uk) joined #gnuenterprise.
reinhard (~rm@N809P002.adsl.highway.telekom.at) joined #gnuenterprise.
ra3vat (ds@ics.elcom.ru) joined #gnuenterprise.
psu (peter@manorcon.demon.co.uk) left irc: Read error: 113 (No route to host)
alexey (alex@ns.techkran.vladimir.ru) left irc: Remote closed the connection
alexey (~Alexey@techkran.vladimir.ru) joined #gnuenterprise.
alexey (Alexey@techkran.vladimir.ru) left irc: "[x]chat"
alexey (~Alexey@techkran.vladimir.ru) joined #gnuenterprise.
Yurik (~yrashk@gw.telcos.net.ua) joined #gnuenterprise.
lekma (~kiki@node1142c.a2000.nl) joined #gnuenterprise.
lekma (~kiki@node1142c.a2000.nl) left #gnuenterprise.
jamest (~jamest@hobbes.math.ksu.edu) joined #gnuenterprise.
jcater (~jason@24.92.70.39) joined #gnuenterprise.
Nick change: mdean -> mdean-[work]
alexey (Alexey@techkran.vladimir.ru) left irc: "[x]chat"
ToyMan (~stuq@c5300-4-ip91.albany.thebiz.net) joined #gnuenterprise.
<dneighbo> ToyMan how was the caribbean?
ICJ (foobar@pD95521BF.dip.t-dialin.net) left irc: Read error: 110 (Connection timed out)
<ToyMan> dneighbo: it was fsckin amazing
<ToyMan> how's it going?
<dneighbo> im tired
Remosi (dahoose@210-86-56-206.jetstart.xtra.co.nz) joined #gnuenterprise.
Isomer (dahoose@210-86-56-189.jetstart.xtra.co.nz) left irc: Read error: 110 (Connection timed out)
drochaid (~drochaid@pc2-kirk2-0-cust175.ren.cable.ntl.com) joined #gnuenterprise.
lekma (~kiki@node1142c.a2000.nl) joined #gnuenterprise.
<dneighbo> hello lekma 
<lekma> hi
<lekma> i still got problems with gnue
<dneighbo> ok, maybe i can help?
<lekma> geas doesn't compile
<lekma> saying that ld dosn't finf lpython2.1
<lekma> i did follow instructions
<lekma> for rh 7.1
<lekma> for rh 7.2
dres (dres@4.18.171.42) left irc: Read error: 104 (Connection reset by peer)
dres (~dres@4.18.171.42) joined #gnuenterprise.
<lekma> any idea?
<lekma> btw is the meeting about geas today?
<lekma> ...
<dneighbo> phone call 
<dneighbo> just a second
<lekma> ok
<lekma> hey it's late!!!
<lekma> i have to pick up my girl friend at the airport
<lekma> see you later (maybe during the geas meeting)
Nick change: lekma -> lekma_afk
<dneighbo> hmm ok
<dneighbo> back from phone
<dneighbo> as for compiling geas
<dneighbo> redhat doesnt by default create an .so for python
<dneighbo> so you either need to compile python on your own to generate one
<dneighbo> or ask a debian friend to mail you theirs :)
<dneighbo> at least this was the case on rh6.2 w/ python 1.5.2
reinhard (rm@N809P002.adsl.highway.telekom.at) left irc: "Never wrestle with a pig. You both get dirty and the pig likes it"
Yurik (yrashk@gw.telcos.net.ua) left irc: "reboot"
Yurik (~yrashk@gw.telcos.net.ua) joined #gnuenterprise.
<dneighbo> would you like to play a game?
neilt (~neilt@66.95.5.110) joined #gnuenterprise.
#gnuenterprise: mode change '+o neilt' by ChanServ!ChanServ@services.
<neilt> list games
<jamest> how about global thermonuclear war?
<neilt> yes Dr. Falken, welcome back
reinhard (~rm@N809P002.adsl.highway.telekom.at) joined #gnuenterprise.
<jamest> so
Yurik (yrashk@gw.telcos.net.ua) left irc: "Client Exiting"
<jamest> is the gang all here?
<dneighbo> close
<reinhard> oh
<reinhard> you were waiting for me?
<dneighbo> might have one or two stragglers yet
<dneighbo> i meant to announce it
<dneighbo> but then thought better to not have tons of people to moderate :)
<reinhard> sorry
<reinhard> i thought it started in an hour
<dneighbo> it probably does
<jamest> probabaly
<dneighbo> i dnot have my email to know the time :)
Action: jamest is behind on mail at gnue.org
<jamest> did we set exact time?
<dneighbo> well we made an exact 4 hour block :)
<dneighbo> and we agreed that we would try to get here in that time
<dneighbo> then meet for an hour or so
<dneighbo> just a four hour block was nice if there was some delay time (since some of us are 'busy') ;)
<dneighbo> its very cool to have a team so spread out
<dneighbo> but it makes meeting scheduling a bit rough unless people are disinterested in sleep :)
jan (jan@dial-213-168-64-57.netcologne.de) joined #gnuenterprise.
psu (psu@manorcon.demon.co.uk) joined #gnuenterprise.
<dneighbo> hi jan
<dneighbo> hi psu
<dneighbo> any one have my email
<dneighbo> as to when the meeting block is?
<jan> hi dneighbo
<dneighbo> jan can you give me a url to your site that has the non donut gnurpc sample of common db access?
<dneighbo> btw: do you have an account on ash?
<dneighbo> if not we should get you one so you can post your samples there
<psu> hi dneighbo
<jan> is there a working cvs tree on bash?
<dneighbo> jan if there is not we can make it so
<jamest> gnue cvs tree is on fsf hardware
<jamest> we got confirmation on your copyright today
<jamest> so you can get account
<jan> dneighbo: ok, i will try to get the code running on ash
<jan> good.
<dneighbo> jamest: i think he meant cvs running on ash
<dneighbo> as in so he could run his samles
<dneighbo> not the 'repository'
<dneighbo> but rather a working copy
<jamest> yes
<dneighbo> but i could be incorrect :)
<dneighbo> we should probably run a 'shared' copy
<dneighbo> as to not eat up too much disk space
<dneighbo> is everyone here?
Action: jan thanks dneighbo to explain what he ment
<dneighbo> if so we should probably go ahead and start
<dneighbo> as teh time was listed as of 30 min ago
<dneighbo> so anyone not here, probably wont be here
<dneighbo> but if some folks are finishing things up or will be available next 30 min or so we can wait for them
<dneighbo> ....
Action: drochaid wonders if he should be here
<dneighbo> depends of if you have some gaelic ale to share ;)
<psu> dneighbo chillywilly seemed pretty sure he couldn't make  it
<drochaid> heh
<drochaid> tha i breagha an-diugh :)
<jamest> i'll be here as best I can
<dneighbo> jamest you know if jcater is around?
<dneighbo> neilt
<dneighbo> reinhard?
<reinhard> i am here
Action: jamest aint telling
<jamest> :)
<jamest> i have no idea if he's here
Action: jamest holds out a doughnut
<jamest> hmmmm
<jan> lol
Action: jcater raises his hand
<jamest> must not be here
<jamest> ah
<jamest> just a little slow
<neilt> dont wait on me
<psu> donut lagged?
<jcater> we need an artist
<jcater> to draw a pot-bellied goat sitting behind a keyboard
<jcater> drinking mountain dew
<jcater> eating donuts
<jcater> w/an empty Dot box by the mouse
<jcater> s/Dot/Dots candy
<jamest> roflmao
<jamest> and put it on a developers page
<jcater> ah
<jcater> and a python reference cracked open on top of the monitor
<dneighbo> hmmm i will see what i can do :)
<dneighbo> ok if we are all here.....
zwiskle (~zwiskle@194-208-138-086.TELE.NET) joined #gnuenterprise.
<dneighbo> i think we need to discuss direction
<dneighbo> perhaps to start with we can offer up some things we have learned
<dneighbo> from teh current geas
<dneighbo> then follow that with some 'base' requirements for the next version
<dneighbo> sort them by priority and necessity
<reinhard> may i propose that we define the goals before the direction
<dneighbo> then get more into what that might really mean
<jcater> don't try to be all things at once... start small and work our way up featurewise?
<dneighbo> reinhard absolutely
<dneighbo> my point was the GOAL of this meeting is to come out with a direction :)
<reinhard> um
<reinhard> sorry i was unprecise
<dneighbo> we start with what is wrong now :)
<reinhard> i meant the goal of geas
<dneighbo> then move into what requirement/goal of geas is
<dneighbo> then talk about it more detail
<dneighbo> and end up with a direction :)
<dneighbo> i had ill family so i apologize i had meant to make a meeting agenda and some purdy pictures for common
<dneighbo> reinhard : so based on your last comment
<jamest> well, the original goal of geas was data access abstraction
<dneighbo> lets maybe briefly say where we see lacking issues in current geas
<dneighbo> and what we want out of new geas
<jamest> then somewhere it became an object server
<jamest> which is probably a good thing :)
dneighbo_ (dneighbo@mail.libertydistribution.com) joined #gnuenterprise.
<reinhard> dneighbo_ don't think you'll get two votes this way!
<jcater> rofl
<dneighbo_> well i think that we can all say perhaps the biggest weakness in GEAS as it sits today
<neilt> just so i can beat my dead horse, the major think lacking in geas is forms support
<dneighbo_> is it evolved w/o direction
<dneighbo_> meaning it was something treshna designed for specific own needs without a direct definition of what it was to be
<dneighbo_> and in andrewm's defense he on more than one occassion expressed frustration with all parties on that note :) as we had different needs than his employer :)
<jamest> i think it lacks (in no order) : forms support, security, performance, stability (much, much better than it was btw)
<dneighbo_> maybe that is all we need to say about it
<dneighbo_> is that we know this time we need to have a 'real' plan of attack
<reinhard> what i see what geas was lacking most at andrew's time was
<dneighbo_> i wasnt looking to lay blame on anyone or anything
<reinhard> * coordination
<jamest> in fact, I'm willing to blame stability on my python/orbit stuff
<reinhard> * goal
<dneighbo_> but rather to try to not fall into same thing
<reinhard> * interest from other team members
<dneighbo_> reinhard i agree 100% on coordination and goal
<dneighbo_> i dont think it was so much a team member thing
<reinhard> in fact i think we were quite unfair against andrewm
<reinhard> we did let him work on geas and nobody said a word
<dneighbo_> i.e. if the team isnt working towards the SAME goal
<reinhard> and after a year's work or so
<dneighbo_> why would they be interested
<reinhard> we told him "oh that's nice but it's not what we want actually"
<reinhard> "why didn't you do y instead of x?"
<reinhard> don't get me wrong
<reinhard> this very much includes myself
<neilt> a good portion of the stuff from treshna has already been pulled out
<dneighbo_> in fairness there reinhard
<dneighbo_> from the beginning i specifically told andru that some things they were doing were not what gnue wanted
<reinhard> ok
<dneighbo_> and on many items andrewm agreed and those items didnt make it into the gnue version of geas
<reinhard> you're right
<dneighbo_> some did, but they knew there was difference of direction, but i think the point is
<dneighbo_> that we need to be
<dneighbo_> a. cohesive
<dneighbo_> b. moving towards a goal
<dneighbo_> and in the past i dont think we were either
<reinhard> ok
<reinhard> let's define the goal
<reinhard> i have this
<reinhard> In 2-tier systems, the application logic lies either in the front end
<reinhard> or in the database (via triggers and stored procedures).
<reinhard> The main purpose of the Application Server is to pull the application
<reinhard> logic out of both of them and serve as a middle layer that abstracts
<reinhard> the logic (called @dfn{business rules}) from the user interface as well
<reinhard> as from the database backend.
<reinhard> --
<dneighbo_> correct
<reinhard> would you agree that this is the main purpose why we have geas?
<reinhard> jamest?
<reinhard> all?
<jamest> yes
<jcater> I would agree 100%
<dneighbo_> 1. Remove methods (logic) from client and database.
<jamest> btw - is someone taking notes?
<dneighbo_> is that a fair summarization of that point?
<dneighbo_> jamest: its logged but im trying to build a doc as we speak
<reinhard> dneighbo_: i will update the whitepaper with results of the discussion
<dneighbo_> Goals
<dneighbo_> -----
<dneighbo_> 1. Remove methods (logic) from client and database.
<dneighbo_> is the start of that
Action: drochaid has full logs on if required for later jamest
<dneighbo_> reinhard: good, but im looking for short bullet point summation that fits on one page
<reinhard> ok
<dneighbo_> that basically lists the goals
<reinhard> just want to make sure that we don't duplicate efforts
<dneighbo_> i will then let the whitepaper get into the details. :)
<jamest> maybe a bit of clarification
<reinhard> i would like to define this point as the "constitution of geas"
<dneighbo_> i.e. its worth more than one person taking notes as it help reaffirm we all hearing the same things :)
<jamest> i still want the 2-tier support so triggers and the like wouldn't be removed from forms
<jamest> but in an OO setting they wouldn't be used
<jamest> is that fair?
<reinhard> yes
<dneighbo_> jamest: well i think even in application server stuff at times you put some things in a client
<reinhard> if it doesn't have too much impact on the rest of the goals
<jcater> this is where I get lost
<dneighbo_> i.e. its not desirable and you dont do it offten but i have seen it
<jan> I agree with jamest. 2- tier is also important
<jcater> I don't see how all logic can be removed from a client and put in an app server
<dneighbo_> reinhard : on goal of gnue is that all tools work alone as well as together
<jcater> I understand that core business rules should be there
<reinhard> ok
<jcater> but the actual act of interfacing with the user, I get lost
<reinhard> may we agree to talk about geas here
<dneighbo_> so in that sense forms in 2 tier mode is kind of necessary to hit our larger goal
<reinhard> not about what forms will do or not do
<dneighbo_> reinhard i agree
<reinhard> but about what geas will do or not do
<jamest> that's fine just wanted to clarify 
<reinhard> ok
<dneighbo_> i think jamest was just clarifying he has no problem agreeing that geas can do methods
<jamest> to answer jcater's questoin
<dneighbo_> just as long as it means its not expected to kill method support in forms :)
<jamest> he's talking about having a post-commit trigger in a form update other tables vs having geas do it
<jamest> dneighbo_: exactly
<jcater> I get that
<reinhard> ok
<jcater> but at times I've thought it was being taken further
<reinhard> now we have
<jcater> but, ok
<reinhard> <dneighbo_> Goals
<reinhard> <dneighbo_> -----
<reinhard> <dneighbo_> 1. Remove methods (logic) from client and database.
<reinhard> this is the main goal
<dneighbo_> 1. Allow methods (logic) to execute on appserver instead of on client and/or database.
<dneighbo_> i changed it slightly
<reinhard> the rest of the discussion is about how we will achieve this
<reinhard> ok?
<jamest> ok
<dneighbo_> as its not a DEFINITIVE remove, but rather allowing it to happen
<dneighbo_> reinhard : im not sure
<dneighbo_> i think there are probably more goals than this
<reinhard> ok
<reinhard> what other goals do we have
<reinhard> you mean the techical goals?
<reinhard> like
<dneighbo_> 2. GEAS can communicate with an number of backend datastorage
<reinhard> @item Stability
<reinhard> @item Portability
<reinhard> @item Maintainability
<reinhard> @item Security
<reinhard> @item Possibility to define business rules, methods and data without having
<reinhard> to be a software expert
<reinhard> @item Performance
<reinhard> @item Runtime configurability
<dneighbo_> 3. GEAS can communitate with front ends via a multitude of different tranportaion methods
<reinhard> are these really goals?
<reinhard> for example i tend to be not sure about 3.
<dneighbo_> um let me add one more then i will explain
<reinhard> i would rather say
<reinhard> our goal is portability
<reinhard> if we need different transport methods to achieve this
<dneighbo_> i would not
<dneighbo_> let me explain
<reinhard> then that will be the way we reach our goal
<reinhard> ok
<dneighbo_> to be PORTABLE means you run on different architectures
<dneighbo_> 2. Communicate with an number of backend datastorage.
<dneighbo_> 3. Communitate with front ends via a multitude of different tranportaion methods.
<dneighbo_> 4. Methods support a number of different languages.
<dneighbo_> if i eliminate 2, 3 and 4
<dneighbo_> i can use jboss right now today and be 'portable'
<dneighbo_> and would say there is no goal in GEAS that isnt being met elsewhere
<dneighbo_> why waste our time writing GEAS
<jamest> i think I see where reinhard is comming from though
<jamest> he's wanting overall goals
<dneighbo_> sigh so am i
<jamest> the big picture
<dneighbo_> if our only goal is #1
<dneighbo_> and portablity
<reinhard> sorry i missed the most important goal
<dneighbo_> please lets end this discussion and use an appserver that is already free software
<dneighbo_> and meets our needs
<reinhard> it must be free
<reinhard> :)
<dneighbo_> jboss is gpl
<dneighbo_> but yes that is a good goal to add. :)
<jamest> how about Security as a goal
<reinhard> was in my list of techical goals
<reinhard> as #4
<reinhard> (priority wise)
<dneighbo_> i guess my point is if our core goal doesnt contain distinctly something that makes us different why bother
<reinhard> yeah
<reinhard> why do we bother?
<dneighbo_> i started to look into using another app server a long while back, but the main reason i saw for fressing forward with geas was multiple language support in methods
<dneighbo_> i did not see any other app server (free or otherwise) have this listed in its 'goals'
<reinhard> i don't care about multiple languages
<reinhard> i will only use one
<reinhard> and so will gnue (the application system)
<dneighbo_> so lets use jboss or enhydra
<reinhard> s/don't care/don't care much/
<reinhard> and i don't care much about different means of transport
<dneighbo_> bigger user base, prexisting code, portable
<reinhard> as i will only use one
<reinhard> surely i bother about database independence
<jcater> I think of that as a cool feature, but if that is all that makes us stand apart, then why bother?
<jcater> ^^ that was re languages, not db independence
<reinhard> i hope this is not shocking
<dneighbo_> well i think the things i see as 'stand outs'
<reinhard> but i followed the mailing lists for hours
<dneighbo_> a. multi language
<dneighbo_> b. multi transport
<reinhard> and about spring 2000 there was an agreement about
<dneighbo_> reinhard the problem is CORBA is dead
<reinhard> * gnue will use python as the application language
<dneighbo_> every body right now is using XML-RPC
<reinhard> * gnue will use corba as transport layer
<dneighbo_> but what will it be next month?
<dneighbo_> SOAP???
<dneighbo_> M$ FOO TRANSPORT?
<reinhard> * gnue will be db independent
<dneighbo_> this is why i am big on transport independence
<reinhard> at least this is what i remember
<dneighbo_> its so immature still there is no good single answer
<dneighbo_> i still agree about python as application language
<jamest> reinhard: that's what I recall too
<dneighbo_> i still agree db independent
sjc__ (~sjc@modem-175.bonobo.dialup.pol.co.uk) joined #gnuenterprise.
<jamest> but i also recall wishing to support more than one transport
<reinhard> ok
<dneighbo_> i think in retrospect that not wanting to support multi transport was short sighted
<reinhard> sorry i don't know either jboss nor the other thing you said
<dneighbo_> 2 years ago i was all for corba only supporting multiple orbs
<dneighbo_> but in playing with a lot of different projects
<dneighbo_> rpc is like database
<dneighbo_> everyone seems to have their own favorite one
<reinhard> if it is true that these other appservers do all what we need except multi transport
<dneighbo_> and only supporting one i think will hurt us long term
<reinhard> then i really ask why write a new one
<dneighbo_> reinhard: i was SERIOUS when i asked that question
<dneighbo_> not making fun or light of the situation
<reinhard> i am serious now, too
<dneighbo_> but frankly saying if we arent doing something different then why reinvent the wheel
<reinhard> until 10 minutes ago i thought there is no free application server at all
<dneighbo_> i have two major reasons i dont like jboss
<jamest> there are quite a few app servers out there
<dneighbo_> a. its not a gnu project (one goal of GNU is to have a full suite of NGU apps)
<dneighbo_> b. its java (as are mostly ALL of the free app servers) and the thought of method code in java hurts me
<jamest> most based upon java IIRC
<jamest> zope might be considered an app server of sorts
<dneighbo_> SO that said that might be the PRIMARY difference
<dneighbo_> is that we would be a NON JAVA app server
<reinhard> ok
<reinhard> so can we add to the list of goas
<reinhard> goals
<reinhard> 2. NOT JAVA
<dneighbo_> jamest: yes zope is considered an app server, but i think that they put everything and hte kitchen sink in zope, so probably not good for this type of application (but i could be wrong)
<jamest> i wouldn't use it like that :)
<jan> I think a very important point is: Possibility to define business rules, methods and data without having to be a software expert (reinhards point 5)
<dneighbo_> jan i think that is probably more implemenation?
<dneighbo_> i dont want people to software experts but what we need to avoid is a CUSTOM gnue language
<dneighbo_> usually people wanting to abstract programming, get into own defined macro language syntax
<dneighbo_> i think a good business rule designer and a relatively clean language like python
<reinhard> please let's not get too much in detail here
Linux (arab@as12-51.qualitynet.net) joined #gnuenterprise.
<reinhard> let's define the goal now
<dneighbo_> should allow a technical person who is not a black belt programer to do well
<reinhard> not the way
<dneighbo_> reinhard yes
<dneighbo_> Goals
<dneighbo_> -----
<dneighbo_> 1. Must be GPL
<dneighbo_> 2. Not Java(tm)
<dneighbo_> 2. Allow methods (logic) to execute on appserver instead of on client and/or database.
<dneighbo_> 3. Communicate with an number of backend datastorage.
<dneighbo_> 4. Communitate via a multitude of different tranportaion methods.
<dneighbo_> 5. Methods support a number of different languages. (lesser goal)
<dneighbo_> is what i have on the list
<dneighbo_> at this point those are in order of importance
<jamest> security?
<reinhard> let me list my techical goals again
<reinhard> * stability
<reinhard> * portability
<reinhard> * maintainability
<reinhard> * security
<reinhard> * define w/o being software export
<reinhard> * performance
<reinhard> * configurability at runtime (w/o restart)
ra3vat (ds@ics.elcom.ru) left irc: "Client Exiting"
<jamest> i like reinhard's goals
<reinhard> dneighbo_ could you please note that i don't agree on goals 4 and 5?
<reinhard> i think we can get a list of
<reinhard> a. goals we all agree on
<reinhard> b. goals we have different opinions
<jamest> with the other goals listed being subgoals of his
<jamest> but then I liked the Final Fantasy movie too 
<reinhard> lol
<dneighbo_> Goals
<dneighbo_> -----
<dneighbo_>  1. Must be GPL
<dneighbo_>  2. Not Java(tm)
<dneighbo_>  3. Allow methods (logic) to execute on appserver instead of on client and/or database.
<dneighbo_>  4. Must be reasonably stable.
<dneighbo_>  5. Must be reasonably secure.
<dneighbo_>  6. Must be maintainable by business class users.
<dneighbo_>  7. Must perform reasonably with large quantities of users and/or data.
<dneighbo_>  8. Must be able to configure without restarting
<dneighbo_>  9. Communicate with an number of backend datastorage.
<dneighbo_> 10. Must run on multiple OSes and Architectures.
<dneighbo_> 11. Communitate via a multitude of different tranportaion methods.
<dneighbo_> 12. Methods support a number of different languages. (lesser goal)
<dneighbo_> are what i have with them in order of importance 
<dneighbo_> i.e. the ones at the bottom are things we are less likely to agree on as well
<reinhard> ok
<reinhard> 4. must be absolutely stable
<dneighbo_> im going with more legal terminology
<dneighbo_> nothing is absolutely stable (as evidenced by the pysche of the developers in this channel) :)
<jamest> 5. must be absolutely secure :)
<reinhard> with 6. i was meaning the code must be maintainable by the coders
<dneighbo_> jamest: if its abosultely secure we are done :)
<dneighbo_> as its not usable :)
<jcater> 6. Must be maintanable by business class users with a clue
<dneighbo_> jcater: i ALMOST added that :)
<reinhard> sorry you misunderstood 6
<dneighbo_> reinahrd i merged your easy to use and easy to maintian
<dneighbo_> as i think the goal should be you need not be a hacker to maintain or use the softwaer
<dneighbo_> BUT i dont expect home users to be able to do so
<dneighbo_> more a business analyst type person
<reinhard> sure
<reinhard> but there is still a different thing in
<jamest> how about this goal
<reinhard> maintaining forms code
<psu> in effect, you want the psus of this world to be able to do it
<jamest> we ait
<jamest> er wait
<dneighbo_> ok point taken let me restructure that
<jamest> 7 is general enough to cover my thoughts
<jamest> psu: man, that's a tall order
<jamest> :)
<psu> i.e. systems accountant, not end user, not jcater clone
<psu> would you see python knowledge as a pre-req for 6.?
<dneighbo_> Goals
<dneighbo_> -----
<dneighbo_>  1. Must be GPL
<dneighbo_>  2. Not Java(tm)
<dneighbo_>  3. Allow methods (logic) to execute on appserver instead of on client and/or database.
<dneighbo_>  4. Must be reasonably stable.
<dneighbo_>  5. Must be reasonably secure.
<dneighbo_>  6. Must be easily maintained.
<dneighbo_>  7. Must be usable by business class users (not just programmers).
<dneighbo_>  8. Must perform reasonably with large quantities of users and/or data.
<dneighbo_>  9. Must be easy to configure and adaptable without downtime.
<dneighbo_> 10. Communicate with an number of backend datastorage.
<dneighbo_> 11. Must run on multiple OSes and Architectures.
<dneighbo_> 12. Communitate via a multitude of different tranportaion methods.
<dneighbo_> 13. Methods support a number of different languages. (lesser goal)
<dneighbo_> reinhard does that cover what you were after? note i canged the configuring a bit, as its more than just a restart
<dneighbo_> it really should be easy to configure etc...
<reinhard> yeah great
<reinhard> just to make sure i am understood
<reinhard> 6. relates to the tool's code
<reinhard> like dbdriver.py or such
<dneighbo_> thats the point :) we all want to make sure we are talking same language :)
<dneighbo_> i think 6 goes for the whole shooting match
<jamest> i see the difference reinhard speaks of
<reinhard> not want to discuss this now but as an example
<dneighbo_> the code should be maintainable as well as the running system :)
<jamest> the geas code base was a bit tough to maintain 
<jamest> the objects defined in geas we not as bad 
<reinhard> i don't think it's maintainable to have 20 dbdrivers for 13 db backends
<reinhard> jamest: yes you got the point i think
<jamest> reinhard: i think we hit a good mix in common....meaning basic support is there and the framework is maintainable
<dneighbo_> so we are agreeing on these goals?
<dneighbo_> is there more?
<reinhard> let me check
<dneighbo_> the only thing i dont see here that was kind of in the other requirements
<dneighbo_> object transparency
<dneighbo_> as the goals sit right now i would assume it up to the developer to handle things as objects or not
<dneighbo_> this is how i believe jboss, websphere and other appservers do it
<dneighbo_> they dont map objects to tables and such
<dneighbo_> they let/force the developers to handle such things
<dneighbo_> im not weighing in right or wrong, or what approach we would take
<reinhard> ok i basically agree with the goals
<dneighbo_> im just stating i dont see that on our goal sheet
<reinhard> however i would make 3. be 1.
<reinhard> and 1. and 2. merge into "Must be really free"
<dneighbo_> my reasoning for the numbering
<reinhard> as "not java" sounds a little childish
<reinhard> :)
<dneighbo_> is if 3 comes before not java, i could argue use a java based one :)
<reinhard> neilt: you here?
<reinhard> you're so silent?
<dneighbo_> Goals
<dneighbo_> -----
<dneighbo_>  1. Must be GPL
<dneighbo_>  2. Not Java(tm)
<dneighbo_>  3. Allow methods (logic) to execute on appserver instead of on client and/or database.
<dneighbo_>  4. Must be reasonably stable.
<dneighbo_>  5. Must be reasonably secure.
<dneighbo_>  6. Must be easily maintained.
<dneighbo_>  7. Must be usable by business class users (not just programmers).
<dneighbo_>  8. Must perform reasonably with large quantities of users and/or data.
<dneighbo_>  9. Must be easy to configure and adaptable without downtime.
<dneighbo_> 10. Communicate with an number of backend datastorage.
<dneighbo_> 11. Must run on multiple OSes and Architectures.
<dneighbo_> 12. Communitate via a multitude of different tranportaion methods.
<dneighbo_> 13. Methods support a number of different languages. (lesser goal)
<dneighbo_> er crap
<dneighbo_>  1. Must be GPL and built with truly free tools.
<dneighbo_>  2. Allow methods (logic) to execute on appserver instead of on client and/or database.
<dneighbo_>  3. Must be reasonably stable.
<dneighbo_> how about that as new 1 2 3 reinhard?
<reinhard> well i would exchange 1 and 2
<reinhard> but i see 2 being the "mission" and the rest being "side rules"
<reinhard> so i'm not sure if 2 belongs into this list at all
<reinhard> but i'm not hot about either way
<reinhard> on priority
<neilt> i'm just the business guy, if you all ever build a tool that I can use, I'll use it, so no comment
<reinhard> i would propose to rise performance in priority
<reinhard> neilt: yeah that's exactly what we are looking for
<reinhard> if a tool fulfills these goals will you be able to use it?
<reinhard> are all these goals important for you?
<neilt> nope
<reinhard> any details?
<reinhard> :
<reinhard> :)
<reinhard> ping
<jan> goal: possibility to change buissness objects without breaking old code. (kind of possibility to archive the old schemata)
<reinhard> jan: correct and i think this is implied in "Must be easy to configure and adaptable without downtime"
<jamest> well
<jamest> in the original gnue docs
<jamest> i wanted to be able to alter objects w/o changing their class name via some type of plugin system
<reinhard> ah yes
<reinhard> the car dealer thing :)
<jamest> yip
<jamest> i don't think any app server today does that
sjc__ (sjc@modem-175.bonobo.dialup.pol.co.uk) left irc: Connection reset by peer
<jan> i think its still important, because in germany the financial administration wants to check old ( up to 10 year) financial transaction. so a kind of versioning system is important.
<reinhard> jamest: could you formulate a goal that fits in our list
<dneighbo_> jan: i think that is a different issue
<dneighbo_> as to me that is a data consistency thing
<dneighbo_> not an object consistency thing
<dneighbo_> i.e. financial restatements
<jamest> lol: off-topic on the majordomo2 mailing list.......
<jamest> I've gotten THREE requests in the last day from people whose subscription
<jamest> requests failed because what they put in the "address" field was
<jamest>         215 Main St.  Springfield, IL 99999
<jamest> Please change the field labels to "E-mail address"... *sigh*
<neilt> if you are going to build something that people will use you need define #7
<jamest> reinhard: i'll try, hang on
<jamest> wow
<jamest> rereading what we have above
<jamest> it never metions objects :)
<jamest> i was going to say goal of 
<neilt> most the rest don't matter, they are techy things
<jamest> dynamically extensible(sp?) business packages
<jamest> neilt: very true
<reinhard> neilt: security, performance, stability?
<jan> authorisation? 
<neilt> of course the code needs to be well written, that is like saying that a car has to be able to roll on flat pavement
<reinhard> i imagine going to a customer that wants to decide whether to use gnue or not
<reinhard> i can say
<reinhard> * it is rock stabe
<reinhard> stable
<reinhard> * it is secure
<reinhard> * ...
<reinhard> neilt: agree somehow that these are basic things
<reinhard> but the question is about priority
<reinhard> would you accept for example performace issues to make configuration easier?
<neilt> so we concentrate on security and never put out an application
<reinhard> would you value stability over security
<reinhard> etc
<reinhard> would you accept a security hole if you win performance
<reinhard> etc
<reinhard> for example i would value stability higher than most of the rest
<reinhard> but i would like to know if others think the same way
<neilt> they dont, hence my trying to stay out of this coversation
<jcater> I would agree stability is important
<jcater> like I said in the beginning, though, I think
<reinhard> neilt: i don
<jcater> don't try to be all things at once... start small and work our way up featurewise?
<reinhard> neilt: i don't want to end up in writing an appserver
<reinhard> that doesn't fit business needs
<reinhard> jcater: sure
<reinhard> jcater: the list is ordered by importance
<reinhard> not by order of implementation
<reinhard> however you can't write a system like old geas
<reinhard> saying "i will care about stability later"
<jcater> actually, that's my problem with the old geas
<reinhard> and then when you have a cool dynamic and intuitive configuration system
<neilt> weill i dont care if it runs on multiple systems
<neilt> i want it to run on one
<reinhard> you say "ok now i make it stable"
<neilt> with an architecture that can be expanded
<reinhard> neilt: on which one?
<reinhard> ;)
<jcater> is it tried to be so much but couldn't do much reliably
Linux (arab@as12-51.qualitynet.net) left irc: 
<neilt> i dont care if it support multiple back ends, i just need one to get started
<neilt> but I want an architecture that others can be added
<dneighbo_> i think things are ordered pretty good
<neilt> etc
<dneighbo_> i put performance after security and maintainance/usability on purpose
<neilt> enough said then i guess
<dneighbo_> as a lightning fast system that isnt maintainable or usable isnt much good
<dneighbo_> i think visual basic, java, python and other intreprted languages
<reinhard> neilt: you would give higher priority to 7?
<dneighbo_> are proof positive businesses gladly will take usability/maintainability over performances in MOST cases
<psu> any perf problems are resolvable medium term by hardware
<dneighbo_> well that is the going theory
<dneighbo_> as resources go, people are expensive hardware is cheap
<reinhard> dneighbo_: did you add the "dynamically extensible(sp?) business packages"?
<reinhard> or (formulated as a goal) maybe
<reinhard> must support seperate treatment of basic installation and extensions
<reinhard> or like that?
<dneighbo_> no i never got a line item on it
<reinhard> jamest: does this formulation fit what you mean?
<dtm> hi all
<jamest> back - reading logs
<jamest> well
<jamest> i guess so
<jamest> i'd be interested in hearing neil's goal list too
<reinhard> me too
chillywilly (~danielb@d53.as28.nwbl0.wi.voyager.net) joined #gnuenterprise.
<chillywilly> how goes it guys?
<reinhard> good
<neilt> jamest: then this is for you
<neilt> we need to expand 7 so I can understand what it means
<neilt> i know you guy are now gun shy of objects
<chillywilly> bah, I have blocks for fonts now, I can't see adman thing your typing
<neilt> but we need a central repository for defining the business objects (or what ever you want to call them)
<chillywilly> there
<neilt> So that business functionality can be defined in one place and show up in all the places it needs to in the enterprise
<chillywilly> fixed it
<chillywilly> I agree neilt :)\
<neilt> we need a way to relate the different business parts
<neilt> to each other
<neilt> etc
[15:09] Last message repeated 1 time(s).
<reinhard> neilt: can you give a better formulation of 7.?
<reinhard> as it actually evolved from what you wrote in your mail
<reinhard> about being able to configure without being a software expert
<reinhard> or like tha
<reinhard> that
mozgabe (~chatzilla@firewall.imaxsales.net) joined #gnuenterprise.
<chillywilly> who has a log of previous crap and when did discussions start?
<jamest> neilt: thanks
<jamest> actually I think objects would be usefull
<jamest> very useful
<jamest> to the point that I put them in the original docs 
<dneighbo_> Goals
<dneighbo_> -----
<dneighbo_>  1. Must be GPL and built with truly free tools.
<dneighbo_>  2. Allow methods (logic) to execute on appserver instead of on client and/or database.
<dneighbo_>  3. Must be reasonably stable.
<dneighbo_>  4. Must be reasonably secure.
<dneighbo_>  5. Must be easily maintained.
<dneighbo_>  6. Must be usable by business class users (not just programmers).
<dneighbo_>  7. Must perform reasonably with large quantities of users and/or data.
<jamest> i'd love to see designer let you build objects
<dneighbo_>  8. Must be easy to configure and adaptable without downtime.
<neilt> reinhard: and we need tools to manipulate this definition before the server even works
<dneighbo_>  9. Communicate with an number of backend datastorage.
<dneighbo_> 10. Must run on multiple OSes and Architectures.
<dneighbo_> 11. Communitate via a multitude of different tranportaion methods.
<jan> the main question is : what function does GEAS should provide to the buissness user? what is the improment instead of just using forms? make a 3tier out of a 2tier is not enough. buissness rules at a save place += access control is important, easy (centralized) maintability is important. working without "problems" (stability, security, performance...)  could be answers to this question
<dneighbo_> 12. Methods support a number of different languages. (lesser goal)
<dneighbo_> is the list
<dneighbo_> i dont see what you are discussing with regard to 7
<neilt> the tools should definitely include validation, lookup, help and definition for the different objects
<jamest> i think designer could be extended to do geas object defs
<jamest> i _think_
<chillywilly> it should be a complete business app IDE right?
<jamest> dneighbo_: his number shifted to 6
<dtm> chillywilly: hi
<chillywilly> sup dtm
<dtm> things
<chillywilly> groovy
<mozgabe> dtm
<chillywilly> ah, having a nice comfortable computer chair does wonders for one's back :)
<neilt> jan: excellent question
<jan> :)
<reinhard> neilt: i'm not that sure
<reinhard> a system where only a programmer can define business objects that run on the appserver is at least usable
<reinhard> a system that can define objects that run nowhere isn't usable
<reinhard> so i would say it is a major point in priority
sjc__ (~sjc@modem-1411.bonobo.dialup.pol.co.uk) joined #gnuenterprise.
sjc__ (sjc@modem-1411.bonobo.dialup.pol.co.uk) left irc: "Client Exiting"
<reinhard> but i wouldn't agree about the order of implementation
<neilt> what i really meant
<neilt> was that developement of tools and server should go in parallel
<reinhard> ok
<neilt> so both are dont at the same time
<reinhard> i agree to that
<neilt> since i am not working on server
<dtm> dneighbo:  you mentioned that there's new functionality in DCL cvs.  explain!
mozgabe (chatzilla@firewall.imaxsales.net) left irc: "ChatZilla 0.8.6 [Mozilla rv:0.9.8+/20020228]"
<neilt> its not that important to me
<dtm> dneighbo:  i didn't get an email notification from sf.net of a checkin
<dneighbo_> dtm in cvs vs last release
<dneighbo_> as teh last release runs off a 'branch' 
<dneighbo_> so 'head' has more features just not yet complete
<reinhard> ok now to get this discussion further
<dtm> dneighbo_:  so should i ignore it for now?
<chillywilly> why don't you fewl take it to #dcl dagnabit :P
Action: chillywilly trout whips them accordingly
<reinhard> dneighbo_: i think we should change 6. to fit neilt
<dneighbo_> what do you want it to say?
<reinhard> and we should maybe extend 8. to include that car dealer idea
<dneighbo_> give me the text
<dneighbo_> and i will add it
<reinhard> yeah that's ok
<reinhard> let me formulate the text
<reinhard> as the only not native english speaker here ;)
<chillywilly> this is the list that dneighbo_ pasted?
<dneighbo_> its just i see this as high level stuff
<dneighbo_> so going into more detail on a line item i think is more in an implemenation doc
<reinhard> no
<dneighbo_> as more could be said in detail about security performance etc
<reinhard> i think what neilt said is high level stuff
<dneighbo_> if you want more items thats cool
<dneighbo_> just give them to me :)
Action: dneighbo_ is trying to imply im not getting it, its not that im being lazy
<chillywilly> he's a big picture kinda guy
lekma_afk (kiki@node1142c.a2000.nl) left irc: 
<chillywilly> I like that
<jan> reinhard: there are other non english speakers out here
<reinhard> jan: sorry
<reinhard> :)
<reinhard> what about merging 6 and 8 into something like
<jan> its just that every sentence with more than one word ;) takes two hours to write, --> makes me feel a bit frustrated :(
Action: chillywilly blames england
<chillywilly> psu: :P
Action: chillywilly goes looking for logs to get upt to speed
<reinhard> must be configurable dynamically, centrally, w/o programming skills, without downtime, and in seperate "layers" for various levels of specialization
<reinhard> i'm sure neilt could find better words
<reinhard> i put all what he said into that word "centrally"
<chillywilly> the end packages will be like that right?
<reinhard> i don't even know if this is a valid english word :)
<reinhard> chillywilly: no geas will be like that
<chillywilly> I mena you don't really expect suits to design business objects do you?
<reinhard> chillywilly: sorry but that's the whole point
<psu> chillywilly: design, yes. program, no
<chillywilly> the system should be configurable and cstomizable as that's what lusers do
<reinhard> chillywilly: we don't talk about lusers here
<chillywilly> it's a joke dude
<reinhard> there is something in between a luser and a hurd hacker ;)
<reinhard> and some of these there in between should be able to adapt gnue to their needs
<reinhard> it's about freedom
<chillywilly> the things that they design should be "high-level" right?
<chillywilly> like business processes, work flow, etc.
<reinhard> the user should be free to adapt the software to their needs (freedom 1 IIRC)
<chillywilly> yes, but not everyone is smart enough to do that
<reinhard> sure
<reinhard> but you should
<chillywilly> and to assume that they will is a bit naieve
<reinhard> a. let those who are smart enough let it do
<reinhard> and
<reinhard> b. make your program in a way that they don't have to be _too_ smart
<chillywilly> yes
<reinhard> and if some business wants the item number to be longint instead of char(35)
<chillywilly> they will designed the business process thought right, as that's what they know
<reinhard> they should be able to do it themselves
<chillywilly> they don't know programming
<jan> but many can use M$ Access f.e.
<reinhard> i think someone that is mentally able to write vb macros should also be able to adapt gnue to his needs
<reinhard> jan: exactly my thoughts, just put a lot shorter :)
<jan> :)
<reinhard> is this discussion still alive?
<reinhard> ping jamest
<reinhard> ping jcater
<reinhard> ping neilt
<reinhard> ping dneighbo_
<chillywilly> GNUe, it's VB for you busness
<chillywilly> er, business
<jcater> well
<jcater> that's better than
<jcater> GNUe, it's VD for your business
<jcater> I suppose
<jcater> :)
<chillywilly> gah
<chillywilly> hello...
<reinhard> chillywilly: funny, but when you came all others went away ;)
<psu> jcater: I assume VD = Visual Dephi ?
<psu> and not anything else?
<jcater> lol
<jcater> well, no
<chillywilly> whatever
<jcater> crude joke
<jcater> VD = venerial (sp?) disease
<chillywilly> it's not like anyone gives a shit about what I have to say anyway ;)
<jan> do we also talk about an implementation plan?  
Action: dtm gives chillywilly a shit
<chillywilly> welcome to my world
<dtm> chillywilly: thanks!
<reinhard> jan: actually that's what i wanted to propose next
<reinhard> after we finally agreed on the goals
<chillywilly> that was a continuation of the prior statement that I made
<chillywilly> not an invitatio to give me shit
<chillywilly> :P
<chillywilly> (no one need that anyway)
<chillywilly> I get it regardless
<dtm> hey is this here that previously schedule gnue staff meeting?
Action: jan is waiting for the next step
<chillywilly> ODMG anyone?
<chillywilly> :)
Action: reinhard is waiting on neilt's and jamest's input on the goals list
<jcater> chillywilly: no thank you
<chillywilly> blah
<reinhard> aaaaalbatros?
<chillywilly> uh?
<psu>  /topic intermission?
<reinhard> psu: :)
<reinhard> and now for something completely different
<chillywilly>  /topic chillywilly gets some trout ready for the next round
<reinhard> an implementation plan
<reinhard> may i propose
<reinhard> a. we implement in python for the time being
<chillywilly> NEVER
<chillywilly> really?
<chillywilly> yay
<jcater> reinhard: one thing I don't agree on is that the tools should be finished at the same time
<jcater> I don't think that's a fair statement
<jcater> but that's just MHO
<chillywilly> what is going to be in python? everything?
<reinhard> b. we implement something that fulfills 1-5 and 7 and make the architecture so that the other goals can be reached afterwards
<reinhard> while we simultanously implement the "object designer" and therefore fulfils 6
<reinhard> and 8
<reinhard> jcater: i understood it should be worked on these _simultanously_
<chillywilly> did someone write this list down?
<reinhard> not that they should release 1.0 at the same day :)
<reinhard> chillywilly: yes derek
<reinhard> and gnuebot
<reinhard> and bigbrother :P
<jamest> back
<chillywilly> yes, and that requires me to dig through some crap
<jamest> I'm ok with the goals
<jcater> chillywilly: they were repeated a half dozen times... shouldn't be difficult :)
<jcater> just search for <dneighbo_> 1. 
<jcater> :)
<chillywilly> I can scroll back up
<reinhard> <reinhard> must be configurable dynamically, centrally, w/o programming skills, without downtime, and in seperate "layers" for various levels of specialization
<reinhard> jamest: is this ok for you?
<jamest> that's a big one :)
<jamest> but yes, it's ok with me
<reinhard> do you see your car dealer stuff in the "seperate layers"?
<jamest> yes
<reinhard> ok
<reinhard> ******************************************
<reinhard> derek please merge 6 and 8 into
<reinhard> must be configurable dynamically, centrally, w/o programming skills, without downtime, and in seperate "layers" for various levels of specialization
<reinhard> ******************************************
<jamest> so, now we talk implementation?
<jamest> or subgoals?
<jamest> or trout?
<reinhard> implementation
<reinhard> lol
<reinhard> i love bash
<reinhard> i just typed
<reinhard> imple<tab>
<jamest> lol, i do that too
<reinhard> <reinhard> may i propose
<reinhard> <reinhard> a. we implement in python for the time being
<chillywilly> :P|
<jamest> i would say we should do in python
<jamest> profile
<reinhard> <reinhard> b. we implement something that fulfills 1-5 and 7 and make the architecture so that the other goals can be reached afterwards
<reinhard> <reinhard> while we simultanously implement the "object designer" and therefore fulfils 6
<reinhard> <reinhard> and 8
<jamest> then fix the bottlenecks
<jamest> fix = recode those sections in C/C++ if the following criteria are met
<reinhard> note that on our list maintainability is higher than performance :)
<jamest>   the problem can not be fixed via code cleanup
<jamest>   the problem is not due to bad design
<chillywilly> agreed
<jamest> that is why i like python approach so much
<jamest> it's lends itself to readability
<jamest> and it's highly productive
<jamest> um
<reinhard> jamest: your opinion about b.?
<chillywilly> doesn't make you design better software though ;)
<chillywilly> no language can
<chillywilly> python is good though :)
<jcater> it can encourage you to, though
<jcater> just look at perl code
<jcater> it encourages poor coding, imho
<jcater> but our meeting isn't about that
<jcater> sorry for digressing
<jamest> I'm ok with b
<jamest> if we do this
<jamest> sigh
<jamest> users
Action: jamest needs to collect thoughts
<jamest> I was talking to neilt other night in IRC
ToyMan (stuq@c5300-4-ip91.albany.thebiz.net) left irc: "Client Exiting"
<jamest> and thought of reusing GTriggerNSObject (or something like that) as part of a core geas object
zwiskle (~zwiskle@194-208-138-086.TELE.NET) left #gnuenterprise.
<chillywilly> jamest: that's a good ide
<chillywilly> er, idea
<chillywilly> imho
<chillywilly> this is what I have been thiking all along that wee one common 'object'
<chillywilly> which supports triggers, etc.
<jamest> it is generic object that can construct a object heirarchy and exposes various methods/attributes for other objects
<chillywilly> and if you want ot do 2-tier then you just subvert the app server
<chillywilly> I like :)
<jcater> subvert?
<chillywilly> you don't use it
<reinhard> jamest: please let's not discuss coding details
<jamest> ok
<jamest> i guess my point is
<reinhard> i would like to discuss a roadmap if that's ok for you
<jamest> I _think_ a huge part of a python based geas resides in common today
<chillywilly> I agree
<reinhard> ok
<chillywilly> db abstraction, rpc abstraction,...
<reinhard> so the first step was to make common usable for geas
<jamest> and since designer can deal w/ objects based upon common's gobj 
<jan> I would like if the programm would be structured like neilts drawing. With simple interfaces between this parts -> easy maintaneablity
<reinhard> jan: that is a must imho for maintainability
<chillywilly> reinhard: aren't we going to be redoing geas?
<reinhard> yes
<chillywilly> ok
<reinhard> sorry let met put it this way<
<chillywilly> just making sure
<jamest> it might make 6 easier
<reinhard> we need to make common code usable for the new geas team
<chillywilly> we write ageas that uses common
<chillywilly> ?
<reinhard> if we can make common usable for geas yes
<jcater> that shouldn't be hard :)
<chillywilly> I think we coud build a python ODMG binding that uses the db drivers
<reinhard> ok
<chillywilly> that would solve one part
<jan> why don't let geas be a superset of common. You can call it per RPC CORBA... or directly use it
<reinhard> jcater: you know what it takes to make it usable?
<chillywilly> er, common db drivers
<reinhard> 1. document interfaces
<chillywilly> well maybe a few more
<chillywilly> than just one part
<reinhard> 2. define stable and unstable parts of the code
<reinhard> 3. define behaviour of the objects
<reinhard> where define == define in write
<reinhard> and define == define once and keep updated
<reinhard> you still sure about
<reinhard> <jcater> that shouldn't be hard :)
<reinhard> ?
<jan> reinhard: I tryed to use the common db drivers for the php client: the only problem seems to be that the objects needs attributs from the gfd file to work
<reinhard> :)
<jcater> given those definitions
<jcater> 'it's probably easier than I girst thought
<jamest> i'm going to have to agree with jcater
<reinhard> coool
<jamest> we haven't done the best job documenting common
<jamest> but she's a beastier
<chillywilly> hrrrm, so gobj needs to be more generic jan?
<jamest> er
<reinhard> however this is what i _now_ see as missing
<jamest> beastie
<reinhard> not saying that i won't come up with more :)
<jamest> gobj is very generic
<jan> not gobj.  gdatasource, gdataobj
<chillywilly> I looked at the code
<jamest> jan: we need to clean that up
<jan> :)
<jamest> if we break encapsulation we should fix it
<jcater> what are we talking about now?
<jamest> that is part of the uidrivers system problem today
<chillywilly> jcater: a python geas that uses common
<dneighbo_> sorry was away
<jamest> ah crap
<dneighbo_> and nearly impossible to read backlog here cause you all keep talking :)
<jamest> masta's back
<chillywilly> ack!
<reinhard> however i think that those documentations are the most urgent part on our way to new geas
<jamest> there goes our productivity
<reinhard> dneighbo_: we decided to release 1.0 of new geas in 2 weeks
<chillywilly> dneighbo_: aren't you using PuTTY?
<jamest> dneighbo_: check your dcl account for your assignments
<chillywilly> reinhard: we did?
<jamest> shouldn't be too hard
<jamest> derek: implement geas based upon common
<jamest> see, only 1 thing in there for you
<reinhard> rofl
<jcater> jamest
<jcater> I thought there was 2
<jcater> 1. document common
<jcater> 2. write geas based on common
<chillywilly> x:)
<dneighbo_> so we decided to do geas on common?
<chillywilly> that's a target on my forehead
<jamest> ah crap, sorry to mislead you derek
<dneighbo_> chilly using putty yes
<jamest> that's 2 items
<chillywilly> well if it's common, then shouldn't it be common to all of GNUe?
<dneighbo_> but in scrolling backwards it pops to bottom every time you talk :)
<chillywilly> aaah
<chillywilly> that blows
<jcater> dneighbo_: read faster
<chillywilly> ;P
<chillywilly> dneighbo_: what about just pulling the log into a browser?
<chillywilly> ;)
<reinhard> let me put here what we actually decided
<jamest> derek: can you ctrl-s the screen
<reinhard> 1. we will do new implementation of geas in python
<jamest> then ctrl-q when done?
<chillywilly> and the crowd goes wild
<reinhard> 2. jamest and jcater will document common
<reinhard> 3. we will look at common and see what we can use there and/or what has to be changed
<chillywilly> I think most if it will be good to use
<dneighbo> read back log
<chillywilly> at leatst db abd rpc
<chillywilly> as when I looked at those they looked pretty generic
<dneighbo> switch to bitchx session
<dneighbo> ok
<dneighbo_> are the goals still valid
<reinhard> 4. we will implement a prototype that is possibly limited
<chillywilly> althought rpc is not finished
<dneighbo_> or shoudl i throw them out?
<chillywilly> reinhard: good plan :)
<reinhard> but the architecture will be extensible.
<dneighbo_> if they are valid is the task list reinhard is spewing real or just teasing me :)
<chillywilly> it's real :)
Action: chillywilly pinches dneighbo_
<reinhard> dneighbo_: re goals please search for "*********"
<dneighbo_> i dont like that suggestion directly
<dneighbo_> but i have pasted it into my file until i have a moment to word smith it into something i like
<dneighbo_> i.e. the concept is ok not sure i like the wording :)
<reinhard> dneighbo_ sorry you missed your chance ;)
<chillywilly> lol
<chillywilly> too late, it has been declared masta-proof
<chillywilly> move along
<chillywilly> ;P
<reinhard> ouch
<jcater> dudes
Action: reinhard just remembers he promised to fix geasv1 within this week
<jcater> I've gotta run
<chillywilly> l8r dude
<reinhard> l8r jcater
<jan> l8r
<chillywilly> heh that ryhmes :)
<chillywilly> rhymes even
<jan> yes and im pumuckl
jcater (~jason@24.92.70.39) left #gnuenterprise ("Client Exiting").
<jamest> dneighbo_?
<jamest> no more comments?
<reinhard> ok
<reinhard> jamest: you will document common?
<reinhard> rsn?
<chillywilly> wtf is rsn?
<reinhard> real soon now
<reinhard> :)
<reinhard> wtf is wtf??
<chillywilly> what the fuck
<reinhard> j/k
<chillywilly> :P
<jamest> i can start documenting it yes
<chillywilly> weeee
<reinhard> cool
<chillywilly> awesome
<reinhard> now the question of the day
<neilt> are we going to discuss the API's before implementation?
<neilt> or just write code
<dneighbo_> im sorry too busy now
<chillywilly> I would prefer a little design :)
<reinhard> neilt: API definition is the first step of implementation
<neilt> that is not the history
<reinhard> upwards api i mean
<chillywilly>  course not :)
<neilt> but i'll wait and see
<reinhard> who will want to discuss the api?
<reinhard> :)
<reinhard> well anyway
<reinhard> now the question of the day
<reinhard> the GEASv2 team consists of:
<reinhard> ?
<chillywilly> afaics we have a design for a bunch of parts in odmg.txt
<chillywilly> API design
<reinhard> any takers?
<chillywilly> me!
<jamest> um
<jamest> we're still using packages right?
<jamest> in the cvs tree?
<reinhard> yes
<jamest> whew
<reinhard> i guess
<jamest> good
<jan> i like to help, but dunno how much time i have...
<chillywilly> well, I would like to use structly ODL syntax
<chillywilly> it has a 'relationship' concept that is pretty cool
<jamest> well, since I'm one of the losers that has been pushing common
<jamest> I'm happy to help
<jamest> and since a recent job site decided to switch from python to java
<jamest> i should have more time to help :)
<chillywilly> :(
<reinhard> jamest: i would like to see your role in
<chillywilly> what did they do that for?
<reinhard> 1. documenting common (interface-like)
<reinhard> 2. keeping the maintainance of common (when we need changes)
<chillywilly>       class Professor
<chillywilly>       {
<chillywilly>         relationship set<Course> teaches
<chillywilly>              inverse Course::is_taught_by;
<chillywilly>       ...
<chillywilly>       };
<chillywilly> class Course
<chillywilly>       {
<chillywilly>         relationship Professor is_taught_by
<chillywilly>              inverse Professor::teaches;
<chillywilly>       ...
<chillywilly>       };
<chillywilly> is the same thing as forign key stuff in 2-tier
<chillywilly> er, foreign
<chillywilly> sorry
<chillywilly> did I cutt you off reinhard?
<chillywilly> I'll stop
<reinhard> no
<reinhard> thanks :)
<jamest> reinhard: that is fine w/ me
<reinhard> cool
<jan> chillywilly: do we need a gfd -> odmg mapper/converter?
<chillywilly> gfd is the UI
<chillywilly> you mean gcd?
<reinhard> but you are aware that you are the one to take the first step on new geas?
<chillywilly> I think gcd should change to ODL
<chillywilly> me?
<reinhard> well of course we could simultanously define the interface
<reinhard> i mean the api geas shows to the front end
<reinhard> ok for the logs
<reinhard> ******************************************
<reinhard> next steps:
<reinhard> jamest/jcater: document common
<reinhard> dneighbo: write down the result of this discussion
<reinhard> reinhard: prepare proposal for GEASv2 API against front end
<reinhard> all: comment on this API
<reinhard> ******************************************
<reinhard> is this ok?
<chillywilly> I want tomsehting to do
<chillywilly> er, something
<chillywilly> pcik me!
[16:34] Last message repeated 1 time(s).
<chillywilly> soh
Action: chillywilly s getting to excited to type
<dneighbo_> reinhard i will email you a dcl account
<dneighbo_> and chilly too
<dneighbo_> and these will go in as tasks
<dneighbo_> i will try to read log later, though i doubt it will happen soon, i do have notes for part i attended however. :)
<dneighbo_> i think this was good discussion 
<dneighbo_> must run now be back later
dneighbo_ (dneighbo@mail.libertydistribution.com) left irc: "BitchX: its what's for lunch"
<psu> I will do a KC special edition asap
<reinhard> psu you are the man
<psu> probably by this time tomorrow nite
<chillywilly> :)
<chillywilly> go psu!
<reinhard> psu: where do you live?
<psu> reinhard: england
<psu> +0000
<reinhard> near london?
<psu> not as late as you
<psu> reinhard: up north (Sunderland)
<jamest> sound good
<psu> almost Scotland, if you believe the southerners ;-)
<reinhard> ok if i ever come close to that point please remember that i owe you a dinner :)
psu (psu@manorcon.demon.co.uk) left #gnuenterprise ("thanks bbl").
<reinhard> i guess the discussion's over
<chillywilly> you may sleep now :)
<jan> looks like
<chillywilly> sweet dreams
<reinhard> for the rest of the week i will spend my gnue time on fixing GEASv1
<reinhard> however i'm not quite sure if that makes sense
<chillywilly> fixing it so that it is BugFree(tm)?
<reinhard> i guess i will put some 1 or 2 hours work into it to make it at least run all demos without errors
<neilt> why?
<reinhard> why what?
<neilt> without forms it is useless
<chillywilly> it's all we got
<neilt> and there will never be forms support for geasV1
<reinhard> neilt: sorry i thought you had the demand to fix GEASv1
<neilt> only because everyone thinks it is broke
<chillywilly> neilt: there's a driver for geasv1
<reinhard> i could leave geasv1 in the current broken state forever without a thought
<neilt> and all of the python API works
<neilt> when the parser is fixed
<reinhard> neilt: what would you think? what should i do?
<neilt> chillywilly: does not work
<neilt> i would not spend a minute on it
<chillywilly> yes, so is it worth fixinbg?
<neilt> the team has spoken
<reinhard> neilt: ok
<jan> btw.  how should forms access GEASv2? should form use a dbdriver again, or should there be some extra abstraction level?
<neilt> there needs to be a network abstraction
<neilt> current geas uses CORBA
<chillywilly> neilt: I thought I was part of the *team*
<reinhard> i guess forms can't use geas via dbdriver because geas isn't a db
<neilt> whitch is too compilcated for the rest of the team
<jan> neilt: i don't speak of network abstraction.
<chillywilly> yes, I always thought it was lame that forms forced geas into a 2-tier model
<neilt> sorry, 
<chillywilly> no offense to anyone else
<chillywilly> just I don't think that works well
<jamest> forms doesn't force 2 tier
<jan> neilt: i think of some object sitting between forms and db abstraction. in a two-tier modell its the trigger engine and some other stuff...
<jamest> the datasource system in forms supports twy systems
<chillywilly> jamest: you data object deal without record and result sets thought right?
<jan> in the 3-tier way its geas who is called per RPC
<chillywilly> that's relational
<chillywilly> ok, I meant to say a relational model
<chillywilly> sorry
Action: chillywilly is screwing up his terms
<reinhard> chchchchrrrrrrr
[16:47] Last message repeated 1 time(s).
<jamest> reinhard?
<chillywilly> uh?
<reinhard> uups
<reinhard> just fell asleep
<chillywilly> hehe
<reinhard> think i should go to bed
<reinhard> ;)
<chillywilly> gog to bed now dude ;(
<jamest> nite
<chillywilly> er, ;)
<neilt> jan: yep
<reinhard> nite all
<chillywilly> night :)
<neilt> nite
<jan> nite
reinhard (rm@N809P002.adsl.highway.telekom.at) left irc: "Friends may come and go, but enemies accumulate"
<jamest> I'm all for a object model in forms
<jamest> it's not fleshed out though
<jamest> jan: what time is it there?
chillywilly_ (~danielb@d53.as28.nwbl0.wi.voyager.net) joined #gnuenterprise.
<jan> 23:49
<jamest> ah
chillywilly (danielb@d53.as28.nwbl0.wi.voyager.net) left irc: Killed (NickServ (Ghost: chillywilly_!~danielb@d53.as28.nwbl0.wi.voyager.net))
<chillywilly_> damn, I keeo kicking the fucking power cord and truning off the box upstairs ;P
Action: chillywilly_ ran downstairs to his laptop :)
Nick change: chillywilly_ -> chillywilly
<jamest> we can get your cvs account setup tomorrow then
<jamest> as I think i have to run^H^H^Hgo to the store after work
<jan> cool.
<jan> whats n^H^H^Hgo?
<chillywilly> a simulation of a badly fongired terminal
<chillywilly> er, configured
<jamest> these guys are mean to me
<chillywilly> :P
<chillywilly> er?
<jamest> when I say I need to run to the store they tell me I should drive
<jamest> but I always type run
<chillywilly> heheh
<chillywilly> I don't do it though
<chillywilly> I blame da masta
neilt (neilt@66.95.5.110) left irc: 
<jamest> l8r
jamest (jamest@hobbes.math.ksu.edu) left irc: "[x]chat"
<jan> n8t
jan (jan@dial-213-168-64-57.netcologne.de) left irc: "KVIrc 2.1.2-pre2 'Monolith'"
chillywilly_ (~danielb@d23.as28.nwbl0.wi.voyager.net) joined #gnuenterprise.
jan (jan@dial-213-168-64-57.netcologne.de) joined #gnuenterprise.
jan (jan@dial-213-168-64-57.netcologne.de) left irc: "KVIrc 2.1.2-pre2 'Monolith'"
chillywilly (danielb@d53.as28.nwbl0.wi.voyager.net) left irc: Read error: 113 (No route to host)
chillywilly (~danielb@d23.as28.nwbl0.wi.voyager.net) joined #gnuenterprise.
<chillywilly> anyome know how to extract a windows .cab file?
<drochaid> use summat like cabviewer
<drochaid> not sure exactly how much control you get tho
<chillywilly> well I want to use wine to do it ;)
fil_c (~username@public1-cosh3-3-cust12.cosh.broadband.ntl.com) joined #gnuenterprise.
fil_c (username@public1-cosh3-3-cust12.cosh.broadband.ntl.com) left irc: Client Quit
jamest (~jamest@fh-dialup-201091.flinthills.com) joined #gnuenterprise.
<drochaid> hey jamest
<jamest> hello
<drochaid> get anything sorted out earlier?
<drochaid> I wasn't able to stick around for most of it
<jamest> yip
<jamest> we're reimplementing geas in python
<jamest> using common
<drochaid> cool
<chillywilly_> damn skippy
<drochaid> that sounds like a lot of work for someone
<drochaid> chillywilly_?
<jamest> i don't think it'll be too bad
<chillywilly_> drochaid?
<drochaid> my business officially launched 33 mins ago ...
Action: chillywilly_ likes the AquaXen theme :)
<drochaid> I'll be designing some business processes and systems for it over the coming months, as I work out what I'm doing
<chillywilly_> woohoo!
<chillywilly_> what business is that drochaid?
<drochaid> and I'll be working with GNUe and postgres as primary systems with any luck
<drochaid> chillywilly_: small systems, internet and backup & disaster recovery consultancy primarily
<chillywilly_> ok
<drochaid> lots of MS type stuffy >:)
<chillywilly_> yuck
<chillywilly_> you got a name and/or web site?
Action: chillywilly_ is thinking of starting something of his own soon
Action: chillywilly_ has a name picked out
<chillywilly_> you got a business plan drochaid?
<drochaid> I do indeed chillywilly_
<chillywilly_> I heard those are pretty important ;)
Action: chillywilly_ pokes dtm
<drochaid> The Morrell Consultancy be the name .... 
<drochaid> web site will be coming shortly
<chillywilly_> cool
<drochaid> http://www.morrell-consultancy.co.uk for a placeholder
<chillywilly_> are you in the US/.
<drochaid> nope, UK
<chillywilly_> ?
<chillywilly_> guess not :P
<chillywilly_> what legal forms do they have in the UK?
<chillywilly_> legal business forms
<drochaid> yeah, business plans are really important .. I believe they're a constantly evolving document, not just something to get you started
<drochaid> sole trader, partnerships, lin=mited liability partnerships, limited liability company and public liability company 
<chillywilly_> yea
<drochaid> only the public liability company is listed on a stock market
<chillywilly_> what form is yours?
<drochaid> initially sole trader
<chillywilly_> sole proprietor, er, "trader"
<chillywilly_> sorry, that's the lamerican term
jcater (~jason@24.92.70.39) joined #gnuenterprise.
jcater_ (~jason@24.92.70.39) joined #gnuenterprise.
<chillywilly_> ;P
<jcater_> ack!
<jcater_> two chillywilly's
<drochaid> I'm starting with no borrowing, so I don't need any protection from creditors, and it's a lot less paperwork at this stage ;)
<jamest> yo
<chillywilly_> yea, that's what I figured would be good to start out as
<drochaid> and two jcater's :o
<chillywilly_> sup dawg
<jcater_> um
<jcater_> why are there two jcaters?
<drochaid> chillywilly_: I'll transfer to a Ltd when I expand
<chillywilly_> I am logged in on the box upstairs
<chillywilly_> and down here too
jcater_ (jason@24.92.70.39) left irc: Client Quit
<jcater> ah
<jcater> 'cause I clicked the icon twice
<drochaid> heh
<chillywilly_> Ltd is which one?
<chillywilly_> jcater: hehe
<chillywilly_> winblows brain fart
Action: chillywilly_ likes single click
<drochaid> Ltd is the limited liabiliy company
<chillywilly_> I have nautilus set to use single click
<chillywilly_> ok
<drochaid> unlisted, full legal protection
<chillywilly_> gotcha
Action: chillywilly_ listens to Rush - The Trees
Action: drochaid uhm, doesn't
<chillywilly_> you don't like Rush?
<drochaid> chillywilly_: what are you thinking of doing?
<chillywilly_> that's a band ya know ;)
<drochaid> I'm not sure if I've heard them before or not
<chillywilly_> GNUe consultancy of course ;)
<drochaid> lol
<drochaid> an area I'll be considering very closely once GNUe has progressed somewhat ;)
<jcater> eh?
<drochaid> lol
<chillywilly_> jcater: he was asking about bidness
<drochaid> uh?
<drochaid> ah
<chillywilly_> you know dawg
<chillywilly_> bidness
<chillywilly_> sheeeet
<drochaid> I know nothing ... that's why I've launched a consultancy ;)
<chillywilly_> :D
<chillywilly_> rock on!
Action: chillywilly_ does some more ODMG notes...
<drochaid> ODMG?
<chillywilly_> Object Data Management Group
<chillywilly_> http://www.odmg.org
<drochaid> ah yeah, them
<chillywilly_> persistent objects dude!
<chillywilly_> we need a python binding now ;)
<drochaid> is the same one that was around in '97?
Action: chillywilly_ is going to name his 'trout'
<chillywilly_> er, I dunno how long they've been around
<chillywilly_> why are all the TLA domains taken :(
<chillywilly_> blah, the net sucks!
<drochaid> that's the last time I looked at them
Action: chillywilly_ is away: dinner
Action: drochaid is away: bed
Action: chillywilly_ is back (gone 00:01:51)
Action: drochaid is still away
<jcater> that was a quick diiner
<jcater> what did you have?
<jcater> a glass of tea?
<chillywilly_> nah
<chillywilly_> I brought it down stairs
<chillywilly_> I just inhaled it
<chillywilly_> it was a grilled chicken sandwhich
<chillywilly_> with mayo, lettuce, tomato
<jcater> mmmm
<jcater> sounds good
<chillywilly_> grilled chicken breast
<chillywilly_> it was good
<chillywilly_> I need another
<dnSleep> jamest: cause im dumb and too lazy to read whole backlog
<dnSleep> there was discussion on how forms would access geas
<dnSleep> is it my misconception or would geas not just be a 'driver'
Nick change: dnSleep -> derek
<jcater> telepathy
<jcater> that would bring a whole new meaning to wireless networking
<derek> telepathy has same issues as 802.11b (not long enough range) :)
<chillywilly_> troutslap(tm) protocol
<derek> hmmm troutslap sounds good, but what happens if it gets fried
<derek> seriously
<jcater> all the better
<derek> how do jcater/jamest see it working
<jcater> as a driver
<chillywilly_> I think it should use the exported gnurpc interface ;)
<jcater> at least, that's my knee-jerk reaction
<jcater> chillywilly_: ya
<jcater> but still via a common driver I'd think
<jcater> that way, reports would use the same "driver"
<jcater> as would integrator, etc
<chillywilly_> I guess so, I dunno nitty gritty detailks of the db drivers
<derek> right, thats how i see it
<derek> but i dont claim to see the internals as you do :)
<jcater> lol
<jcater> right?
<jcater> so there was a right and wrong answer?
<jcater> :)
<chillywilly_> does you common db layer make it as a relational source with records sets?
<jcater> to a certain extent
<jcater> but that's not set in stone
<jcater> i.e., at that level, it's more terminology than anything
<jcater> but, we'll see
<chillywilly_> ok
<jcater> we've put 0 effort into this new geas so far
<derek> im excited
<chillywilly_> just wondering if there should be an API for a relational source and a different one for objects
<dtm> yeah, dan, get on the stick.  do what jcater says.
<chillywilly_> but I dunno
<dtm> derek: hey thanks for that mp3   it is playing now
<chillywilly_> so you have pluggable db backends abd pluggable object backends, so to speak
<chillywilly_> I dunno if there's much of a difference and I haven't given it much thought
<chillywilly_> derek: me too :)
<chillywilly_> dtm: yessir!
<dtm> dubba da dup dup da doo!!
<chillywilly_> jcater: that's not true, I have been documenting the ODMG standard that's at least somehwat new geas relevant ;)
<jcater> we're doing objects?
<jcater> I wasn't sure if we were starting out with that or not
<chillywilly_> well, I would like to use inheritance and such
<chillywilly_> the ODMG Object Model is pretty simplifies compared to what andrewm wanted to do
<chillywilly_> er, simplified
<jcater> I'll be honest
<jcater> I've had 0 time to look at your notes
<chillywilly_> :)
<chillywilly_> as has everyone else
<chillywilly_> oh well
<chillywilly_> it's there
<chillywilly_> so I can say
<chillywilly_> RFTN
<chillywilly_> er, RTFN
<chillywilly_> j/k
Action: drochaid is away: no really, this time I am ...
<drochaid> night doods
<chillywilly_> night
<chillywilly_> well, a nice integrated GNUe would sure be nice :)
<chillywilly_> all parts working together harmoniously
<derek> i havent read them really
<chillywilly_> this is the vibe I got from thing anyway
<chillywilly_> things
<derek> which would be better...
<chillywilly_> derek: no biggie, maybe you will someday, maybe you won't
<derek> cvs.gnuenterprise.org
<derek> or
<derek> www.gnuenterprise.org/cvs/
<derek> as i hate having to go to the home page to find our cvs tree online
<chillywilly_> first one
<chillywilly_> imho
<derek> cvs.gnue.org used to work, but its horribly out of rsync i fear
<chillywilly_> hehe, rsync
<chillywilly_> is this to mirror the repository or just that web access stuff?
derek (derek@cpe-24-221-112-50.az.sprintbbd.net) left irc: Remote closed the connection
derek (~derek@cpe-24-221-112-50.az.sprintbbd.net) joined #gnuenterprise.
<chillywilly_> wb derek
<derek> that was wild
<derek> X got REALLY slow
<derek> was jsut about to bitch about it in here
<derek> and X crashed
<chillywilly_> lol, the masta curse
<derek> i hope its not a trend 
<derek> i had a lot of stuff open for more than a week
<derek> my guess is something has a leak
<chillywilly_> yea
<derek> probably mozilla mail client
<derek> sigh
<chillywilly_> use mutt
<derek> i need to get a debian box NOW
<chillywilly_> :P
<chillywilly_> aaah, deadcat
<chillywilly_> poor masta
<dtm> derek: what's this with you and demanding a certain distro? :)
Action: derek is currently listening to : 14-15=( 09-JOSE_FELICIANO-Feliz_Navidad.ogg 15| 2:59 )15=14-
<derek> dtm: life is hell w/o debian
<derek> not as bad as windows, but evil
<chillywilly_> cause debian rocks yo socks dtm
<chillywilly_> however
<chillywilly_> installing debian can be like pulling teeth
<dtm> what in the world is so horrible without debian?
<derek> honestly if redhat could update me from 6.2 to 7.2 with a few commmand lines
<derek> i wouldnt be in a hurry to get rid of redhat on this box
<dtm> derek: are you talking about including the kernel and glibc?
<dtm> derek: in the update?
<derek> as i like how this machine works and i dont want to 'disturb' that
<derek> dtm: yes
<derek> but im relegated to installing everything from source now
<derek> and its getting a bitch to maintain
<dtm> derek: are you installing stuff that doesn't come with rh7.2 at all?
<chillywilly_> just run debian unstable and monitor things before you do an upgrade and life is peachy :P
<dtm> derek: i'm doing everything from SRPMs myself
<derek> dtm: dont know as i dont even look for rpms anymore
<derek> sadly of late i have been living of tar.gz
<dtm> derek: well if you're installing common things that come with rh7.2 then you need only issue 'rpm --rebuild *src.rpm' and then you'll have everything
<jcater> I only have one non-Debian linux machine at work now
<jcater> yay
Action: derek is currently listening to : 14-15=( METALLICA-For_Whom_The_Bell_Tolls 15| 5:05 )15=14-
<dtm> derek: once, i updated from rh 6.0 to 6.2 with a cdrom and a reboot
<dtm> derek: everything was completely in place
<chillywilly_> brb
<dtm> derek: before that, i updated from rh4.2 (libc5) to rh5.1 (libc6) runtime in one single command
<derek> hmmm
<dtm> derek: that is, i did one 'rpm' command which installed libc6 in addition to libc5 and everything else needed to run gimp.
<derek> i really want debian :)
<derek> all my friends use debian, i have to use debian too
<derek> debian makes you cool
<dtm> derek: what type of stuff are you trying to install?
<dtm> derek: yeah that's what i hear
<jcater> my redhat upgrades were always hell :)
<jcater> :(
<dtm> derek: and everyone i know, loves debian because it's incredibly difficult to realize how easy it is.
<dtm> derek: so they not only have a tremendous and real comfort zone with it, they have an elite club
<dtm> derek: it's truly unix-like
<chillywilly_> jcater: I agree upgrading redhat never worked out for me either
<dtm> derek: however i've never heard of a single real unique advantage offered by debian other than a larger quantity of potentially available mainstream packages and some extra idealism
<derek> dtm: wow
<chillywilly_> dtm: umm, that *idealism* is one of the reasons we're sitting in this channel ;)
<dtm> how exactly does updating redhat not work?  i'm sure it's possible but my extensive experience has not shown me that.
<derek> i will say debian is much more UNFRIENDLY to install and get running
<jcater> dtm: whoa, buddy
<derek> but once you learn apt
<derek> its not touchable IMHO
<derek> for maintenance
<jcater> I seemlessly upgrade machines right and left now
<jcater> It was hell with many redhat machines
<jcater> oh, and slackware
Action: jcater cringes
Action: jcater will try to avoid ranting
<dtm> derek: yeah like i said though .. "unique" advantage.  i see absolutely no advantage over 'rpm' and 'rpmfind' except often some better made packages which is a community feature, not a tech feature :)
<jcater> dtm: apt goes way beyond rpm
<chillywilly_> not to mention you can downgrade packages wen they are busted
<jcater> when I had a redhat machine
<jcater> I gave up on rpm's
<jcater> and did source installs all the time
<chillywilly_> rpmfind sucks
<chillywilly_> it never even worked half the time I used it
<jcater> as it was impossible to keep up the requirements for rpms
<chillywilly_> this is going back some days though, last deadcat I ran was 5.2 or something
<dtm> derek: and last i knew, .deb didn't have a very good src maintenance infrastructure, unlike rpm.  rpm does maintain multiple patches and has an indepth .spec file.
<derek> whoa?
<chillywilly_> uh, no
<dtm> chillywilly_: i'm talking about what a system _can_ do, not what it's configured by default to do.  otherwise all of em suck  ;)  rpmfind is great if it's configured properly.
<chillywilly_> .debs manage pacthes just fine
<dtm> derek: so educate me if you like
<derek> apt is 100000 times more efficient than rpm + rpmfind
<chillywilly_> apt-get update
<derek> now if you want to argue its not much better than redcarpet
<chillywilly_> apt-get upgrade
<dtm> derek: do you know exactly why though?  so i can get a grip on it?
<chillywilly_> yay!
<derek> you might have more of an arguement
<dtm> derek: ok gotcha
<dtm> derek: that could be
<derek> but then you have the issue of redcarpet is LIMITED severely in what it actual is in charge of 'upgrading'
<derek> put simply... i fell in love with apt in matter of about 30 minutes
<dtm> I have all the anecdotal evidence in the world, that maintaining redhat is extremely simple and easy in every dimension of usage.  And that evidence seems to be completely contrary to what some others would say.  :)
<derek> and generally i dont like things that are command line (i.e. to suggest to other people) i want something my wife would use
<derek> but where debian and apt made me literally  cream my self
Action: derek is currently listening to : 14-15=( VANDALS-Jackass 15| 3:20 )15=14-
<dtm> ok but can you guys tell me the exact specific features and reasons that make debian's subsystems better to you?
<derek> is i needed a newer kernel for a feature of a card
<derek> i went out and there were like hundreds of precompiled kernels for me
<derek> one even for my specific CPU
<chillywilly_> how about I can roll my own kernel debs with kernel-package
<derek> it took about 10 minutes to install it and be on the road
<dtm> derek: from where?  from the canonical debian sites?
<derek> and 9 of that was waiting for jcater/mdean to answer a lilo.conf question
<derek> dtm: all i did was
<derek> apt-get update
<derek> apt-cache search kernel
<derek> apt-cache search kernel |less
<dtm> derek: rpm and rpmfind could do the same thing if the packages were made, and if rpmfind was configured to search the sites containing them
<derek> er
<derek> apt-cache search kernel |grep 2.4|less
<derek> or something
<derek> and then apt-get install thatnewkernel
<dtm> neato
<derek> yes but i didnt have to open a browser go to rpm find
<derek> etc etc etc etc
<derek> this is REALLY evident in DEPENDENCIES
<dtm> derek: nobody opens browsers to go to rpmfind.  it's a cli util :)
<derek> example if i want kde
<derek> i do
<derek> apt-get install koffice kde
<derek> and it gets it all for me pretty much
<derek> er apt-get install koffice kde konq
<dtm> derek: i would go to rpmfind.net if my local rpmfind wasn't configured to search very widely which is usually the case  its default search sites are pretty brane-ded.
Action: derek is currently listening to : 14-15=( TEMPLE_OF_THE_DOG-Hunger_Strike 15| 4:06 )15=14-
<derek> we can agree to disagree
<derek> as at this point at least with rh6.2 you cant convince me its as easy to maintain as debian
<dtm> i'm not disagreeing with anything -- i'm asking questions :)
<chillywilly_> hey I switched to debian because I was sick of buygin redhat CDs and having to do a clean reinstall because the upgrade would never work out right
<derek> what sold me on apt the first time was apache and php
<derek> apt-get install apache
<derek> apt-get install php4
<dtm> i'm only asking if debian's perceived advantages are due mostly or all to its default config, like the list of sites which 'dselect' and 'apt-cache search' are configured to search.
<derek> apt-get install php4-pgsql
<derek> apt-get install php4-xml
<derek> etc etc
<dtm> or the availability of more esoteric combinations of packages like your example with the special kernel
<derek> and boom no editing files or anything no compiling or anything
<chillywilly_> dtm: I had a proper rpmfind config
<derek> i was like WOW
<dtm> derek: yeah same here -- those all come with rh7.2
<chillywilly_> it still worked like crap
<dtm> derek: I have never compiled a single portion of anything apache related.
<derek> even in rh6.2 and rh5.2?
<dtm> derek: that includes support for mysql, pgsql, ssl, php3, php4, etc
<dtm> derek: definately on rh6.2 and up.  i didn't use rh5.2 for that
<derek> er rh5.2, 6.0, 6.2 all made me compile that stuff
<dtm> derek: i think i totally skipped 5.0-5.2
<derek> i still have a rh5.2 box that runs (alpha processor)
<dtm> derek: i was still using 4.2 during that phase because rh's dot-oh releases tend to be insane
<chillywilly_> debian has more packages anyway
<chillywilly_> 6000+ packages
<chillywilly_> it's built by the community
<chillywilly_> it has Free Software Guidelines
<derek> to me the difference is integrity in the system
<chillywilly_> debian seems to be rock solid and pacgaes get updated daily
<derek> red hat doesnt make all the packages for redhat (or at least not the most current ones)
<derek> where as debian as a whole is tested for dependencies etc
Action: derek is currently listening to : 14-15=( STAIND-Crawl 15| 4:29 )15=14-
<derek> i.e. i dont think i have ever gone out and dowloaded a .deb
<chillywilly_> debian has better dependency checking
<chillywilly_> imho
<derek> i am always able to get from REAL debian
<chillywilly_> derek: :)
<derek> i cant say that for red hat
<derek> say i get a rh7.2 disk and in 6 months i need new packages
<chillywilly_> exactly, I was so sick of grabbing an rpm by hand and then having to get get more
<derek> im getting them from some maintainer that doesnt take into account all the other packages on my system
<chillywilly_> basically be my own package manager
<derek> evidence of this in gnue
<derek> is pyXML and xslt4thought or something fight
<derek> if you upgrade pyXML to newest version xslt4suite wont let it as it depends on the old version
<derek> EVERY single rh7.2 gnue install from rpm has asked that question here
<derek> in debian that doesnt happen because the maintainer of xslt or pyxml would be force to fix the package before it would go in
<derek> i hope that makes sense
<chillywilly_> yup
<derek> so even if rpm fixed its 'package distribution' and offerred something more apt-get like
<derek> (i think red carpet is close)
<chillywilly_> higher quality packages, easily upgradable, with the ethical ramifications to boot
<derek> unless they get the breadth of packages (red carpet lacks here)
<derek> and the testing for dependencies (what disparate red hat packaging out side official release does)
<derek> it wont touch debian imho
<derek> i.e. i think redcarpet does the package testing like debian does
<derek> but they are probably closer to a few hundred packages instead of over 5000
<chillywilly_> 6000 you mean ;)
<chillywilly_> probably eve more
<chillywilly_> even
jcater (jason@24.92.70.39) left irc: Connection timed out
<derek> that said i dont bag on red hat
<derek>  i think they are the number 2 distro
<derek> and i think they have put a LOT into the community
<chillywilly_> well they're not s bad company per se
<derek> and i dont think that should go unrecognized
<chillywilly_> s/s/a
<chillywilly_> they just have an ok distro
<chillywilly_> ;)
Action: derek is currently listening to : 14-15=(  -  15| 3:56 )15=14-
<chillywilly_> heh, the untitled song
<chillywilly_> dude where'd you get that script at?
<chillywilly_> (i know I probably asked you this before)
<derek> been so long i dont remember
<derek> its a media plugin for xchat
<chillywilly_> heheh
<derek> combined with an xmms plugin
<chillywilly_> ok
<chillywilly_> it's probably in debian ;)
<derek> hmm this untiled is old nirvana
<derek> from bleach
<chillywilly_> I have Rush - The Trees that does have a proper mp3 header so xmms displays it with no title
<chillywilly_> hmmm, dude did you know they have a console xchat?
<chillywilly_> xchat-text - IRC client for console similar to AmIRC
<chillywilly_> you should try that out when you're stuck on winders
Action: derek is currently listening to : 14-15=( RED_HOT_CHILI_PEPPERS-Sir_Pyscho_Sexy 15| 8:16 )15=14-
<derek> hmmm yes i should
<chillywilly_> the package is only 87.2KB
Action: chillywilly_ downloads it
<derek> tap tap tap
<derek> is this thing on?
<chillywilly> yes?
<derek> ok cool
<chillywilly> whatcha want dawg
<derek> was having problems yesterday thought they returned :)
<jamest> whoa
<derek> whoa?
<jamest> activity
<derek> activity?
<jamest> in here
Action: jamest had window covered up
<derek> ah
<chillywilly> you shouldn't be covering up your windows dude
<chillywilly> ppl want to see in ;P
<nickr> only chillywilly
<nickr> freakin peeper
<chillywilly> damn skippy
<chillywilly> you got a problem with that?
Action: chillywilly wields trout
Action: derek is currently listening to : 14-15=( Sublime - Santeria 15| 3:03 )15=14-
<dtm> chillywilly: hahaha
<chillywilly> I need to get sound, printer, scanner, and cd burner working on this box unde GNU/Linux
<chillywilly> *sigh*
<jamest> bwahahahahahahaha
<chillywilly> but I wan tto work on GNUe!
<chillywilly> it's not fair
<nickr> they also have an X version of irssi
<derek> jamest: i updated cvs yesterday or so
<derek> dneighbo@latitude:~/cvs/hcs/www/forms$ DB000: [DBdriver:350] Database driver needs updated to not initialize DBSIG_DataObject_Object
<derek> is still getting shown
<derek> gnue over ssh tunnel just rocks my socks
<jamest> is this from an setup.py installed copy?
<derek> um no
<derek> gfcvs
<derek> hey i was talking to jcater about case insensitive searching
<derek> did you implement that ?
<derek> i.e. if i do 
<derek> %smurf%
<derek> it wont find John Smurf
Nick change: mdean-[work] -> mdean-[rtcw]
<chillywilly> groovy
<chillywilly> http://hpinkjet.sourceforge.net
<jamest> derek: not implemented yet
<jamest> please tell me your provider
<derek> [hcs]
<derek> comment = HCS Production
<derek> provider = postgresql
<derek> host = localhost:9999
<derek> #port = 9999
<derek> dbname = hcs_prod
<chillywilly> darn no debian pakcgae of the hpijs stuff
<jamest> ok
<jamest> i'll check out the postgresql driver
<chillywilly> CUPS uses ghostscript doesn't it?
<chillywilly> anyone home
<chillywilly> ?
<jamest> ooooooo
<jamest> derek: you here?
<jamest> chillywilly: yip
<derek> im here
<jamest> you notice the new designer wizards?
<derek> no im using production stuff man :)
<derek> you know those elusive apps forms cant do :)
<chillywilly> jamest: how do you use the cups web config interface thingy?
Action: chillywilly cannot remember
<derek> i made them a while back
<jamest> chillywilly: never used it
<jamest> derek: when this commit is finished
<jamest> you can build forms with X number of blocks and various master/detail combos
<jamest> all from a wizard
<derek> weee
<chillywilly> jamest: what do you use?
<jamest> pwd
<derek> so i should test this weekend and we can release sunday night?
<jamest> should be commited
Action: derek is currently listening to : 14-15=( MOTLEY_CRUE-Wildside 15| 4:42 )15=14-
<jamest> lprng w/ apsfilter
<jamest> bbi45min
<chillywilly> oh, you don't use cups
<jamest> nope
<chillywilly> I think I get it to work with the cupsomatic filter generator
<chillywilly> anyone know what happened to ajmichie?
<derek> finals?
<chillywilly> hmmmm
<chillywilly> even then he's usually on IRC ;)
<chillywilly> I think it is summer in NZ right now
<chillywilly> and there is no school
<derek> hmmm maybe home with parents?
<chillywilly> well yes he was at home...maybe he just can't hang out on the computer and use up all the internet minutes 
<chillywilly> (they have a minimal account with a time limit on it)
jcater (~jason@24.92.70.39) joined #gnuenterprise.
<jcater> 'sup?
<chillywilly> whaddup nicka
<derek> sup g
<derek> must run for a bit be back in a few hours
Remosi (dahoose@210-86-56-206.jetstart.xtra.co.nz) left irc: Excess Flood
Remosi (dahoose@210-86-56-206.jetstart.xtra.co.nz) joined #gnuenterprise.
<chillywilly> anyone know how I can create a USB printer device?
<nickr> chillywilly: usbmgr does a lot of that sort of thing for you
<chillywilly> I'm thinking there's some more stuff i need to istall
<nickr> generally devfs magically creates the devices
<nickr> you can cheat and mount devfs somewhere under /dev like /dev/devfs/
<nickr> so you don't have to actually switch
<chillywilly> yea
<chillywilly> I usually use devfs
<chillywilly> BUT
<chillywilly> I just installed this system
<chillywilly> haven't recompiled the kenr elk yet ;)
<chillywilly> er, kernel
<nickr> uh
<nickr> the stock debian kernel has devfs
<nickr> in the 2.4.x series at least
<chillywilly> hmmm
<chillywilly> well it doesn't seem to be running though
<chillywilly> Linux babers 2.4.17-bf2.4 #1 Sun Feb 3 19:56:58 CET 2002 i686 unknown
<chillywilly> (time is wrong :P)
<chillywilly> date is too
<nickr> ave you tried mounting it?
<chillywilly> won't usbmgr make the devices for me?
<chillywilly> nope
<nickr> yea
<nickr> sorta
<chillywilly> Setting up hotplug-utils (0.0.20020114-5) ...
<chillywilly> Setting up usbutils (0.9-1) ...
<chillywilly> Setting up usbview (1.0-1) ...
<chillywilly> Setting up usbmgr (0.4.8-5) ...
<chillywilly> Starting usbmgr: usbmgr.
<chillywilly> jsut installed it
<nickr> I haven't delved deeply into it, but it seems to do a lot
<nickr> I suggest a trip to rtfm
<chillywilly> well I prefer to just give it a shot before I rtfm
<chillywilly> ;)
<nickr> wot? wot? aw bloody'ell
<chillywilly> doh
<chillywilly> ah well
<chillywilly> Feb 28 15:04:31 babers kernel: printer.c: usblp0: USB Bidirectional printer dev 3 if 0 alt 1
<chillywilly> I should get a /dev/usblp0 right?
<chillywilly> er, if I got it going tright
<chillywilly> devfsd doesn't seem to want to run
<chillywilly> I could swear that cups let me pick a USB printer before
<nickr> chillywilly: devfsd won't run if devfs isn't mounted
<chillywilly> oh yea
<chillywilly> DUH
<chillywilly> dude I seriously need sleep
<nickr> yea
<nickr> maybe you should get some
<nickr> sleep that is
<chillywilly> but I want to get this printer working so I feel like I accomplished something :P
<chillywilly> woohoo printer works!
<chillywilly> time for bed
<chillywilly> man the marginsd suck though
<chillywilly> er, margind
<chillywilly> margins
<chillywilly> bah
<chillywilly> sleep(3000);
chillywilly (danielb@d23.as28.nwbl0.wi.voyager.net) left irc: 
alexey (~Alexey@techkran.vladimir.ru) joined #gnuenterprise.
Nick change: chillywilly_ -> chillywilly
chillywilly (danielb@d23.as28.nwbl0.wi.voyager.net) left irc: "sleep(3000);"
jcater (jason@24.92.70.39) left irc: "nite"
drochaid (drochaid@pc2-kirk2-0-cust175.ren.cable.ntl.com) left irc: Read error: 104 (Connection reset by peer)
drochaid_ (~drochaid@pc2-kirk2-0-cust175.ren.cable.ntl.com) joined #gnuenterprise.
Nick change: mdean-[rtcw] -> mdean
--- Fri Mar  1 2002