[00:00] Last message repeated 1 time(s).
<jamest> :)
<jcater> or, during the cheerleading camp... "jamest, take a breath... breathe man... come on"
<derek> jamest: i subsituted 'lusers' for the real persons name to preserve the guilty :)
<jamest> lol
<derek> jcater: no no no
<derek> during cheer camp
<derek> its ''jamest, blink man... blink for heavesw sake... blink''
<jamest> ha ha
<Mr_You> hehe
[00:02] Last message repeated 1 time(s).
<jcater> james is gonna hate me
<Mr_You> see, I like good comedy
<derek> comedy?
<derek> this is biography
<Mr_You> I'm gonna less freakin some channels cause those jokers just become annoying..
<Mr_You> haha
<Mr_You> s/freakin/frequent/
<jcater> we are releasing on a Tue night/Wed morning and I will be gone Thursday morning thru Monday morning... leaving jamest here for support for the hoards of people who come to check us out
<derek> rofl
<jcater> we should just train gnuebot
<jcater> "Did you modify your connections.conf file?"
<Mr_You> yeah I think I've got stuff to do too 
<derek> hmmm my modem seems to be having issues...
<Mr_You> ;-)
<Mr_You> right, me, unemployed.. 
Action: derek gets off phone, mysteriously my providers says there will be an outage for tue night through monday morning
<jcater> derek: you switched to cable modem??
<jcater> :)
<jamest> grrrr
<Mr_You> an example of a breathing technique would be two breaths every minute
<jamest> don't make me come down there derek
<Mr_You> if yer good
<jamest> any show stoppers?
<jamest> jcater?
<derek> jcater: on serious note, next week can we finalize the press releaes for your call center :)
<Mr_You> no curses :-(
<derek> no curses is a result of good breathing technique ;)
<jamest> Mr_You: all pending on this release
<jcater> derek: can try
<Mr_You> heh
<jamest> are we golden?
<jcater> jamest: no show stoppers that I know of
<jamest> i'm still downloading btw
<jcater> ...as long as they don't try to run anything
<derek> Mr_You: my wife told me to please stop cursing....
<jamest> can you check in your inno changes?
<jamest> so I can tag cvs
<jcater> done
<jcater> and done
<Mr_You> some philosophy gives the turtle as a relation to breath and age..
<jamest> sigh
<jamest> modems suck
<Mr_You> they have two way satelite now
<jcater> yip
<jcater> I think my mom's gonna do that
<jcater> 2-way satellite that is
<Mr_You> high latency tho :-( 
<Mr_You> can't do games or voice over IP
<jcater> well, she lives out on a farm
<jcater> they don't even have a post office
<Mr_You> yeah, I'll probably have it one day.. 
<jcater> they have to borrow another town's post office :)
<jcater> hehe
[00:12] Last message repeated 1 time(s).
<jcater> my Intro form is fun to play with
Action: jcater is cracking up over the fortunes
<jcater> it's really, really bad when you laugh at your own jokes
<jamest> :)
<jamest> gnue-common, gnuef, and designer are now tagged 0-1-0
<jamest> it's official 0.1.0 is out
<jcater> woo hoo
<jcater> let the breakage begin
<jamest> um
<jamest> first we have to make the stuff available for download :)
<jcater> damn
Action: jcater will wait for the commits
<jamest> commits?
<jcater> j/k
<Mr_You> hehe
<Mr_You> whooooooooooo hooooooooooooooo!
<jcater> yes?
<Mr_You> 0.1.0
<jcater> this is a big milestone for us
<jcater> this release was 4 months in the making :)
<jamest> :)
<Mr_You> yah ;-)
<jcater> lots and lots of changes :)
<Mr_You> I gotta bring up some windows to show the roommates so they don't think I'm crazy
<jamest> jcater you posting on the site?
Manaic-yawa (darryl@h24-82-132-209.wp.shawcable.net) got netsplit.
nickr (nick@e-172-IP26.empnet.net) got netsplit.
derek (derek@cpe-24-221-112-50.az.sprintbbd.net) got netsplit.
mcb30 (mcb30@pc1-papw2-0-cust211.cam.cable.ntl.com) got netsplit.
<jamest> the annoucement
<jcater> now?
<jcater> lemme reboot into linux
jcater (jason@HubA-mcr-24-165-193-222.midsouth.rr.com) left irc: Read error: 54 (Connection reset by peer)
<jamest> i'm adding links as we speak
nickr (nick@e-172-IP26.empnet.net) returned to #gnuenterprise.
jcater (~jason@HubA-mcr-24-165-193-222.midsouth.rr.com) joined #gnuenterprise.
dres (dres@4.18.171.42) left irc: "see ya"
Manaic-yawa (~darryl@h24-82-132-209.wp.shawcable.net) returned to #gnuenterprise.
derek (~derek@cpe-24-221-112-50.az.sprintbbd.net) returned to #gnuenterprise.
mcb30 (~mcb30@pc1-papw2-0-cust211.cam.cable.ntl.com) returned to #gnuenterprise.
<jcater> ok
<jcater> story posted on home page
<jcater> story=release notice
<jamest> cool
<jamest> i'm about down w/ downloads sectoin if anyone cares to test links
<Mr_You> hmm found a weird bug already ;-)
<Mr_You> File->Open in Designer
<Mr_You> work around: File->New .. then File->Open will work in the new window
<jcater> sigh
<Mr_You> well seemed to anyway
<Mr_You> sorry
<jamest> we have any reason to keep old downloads on downloads page
<jamest> can we just keep the files and say last rev in addition to the current
<jamest> or leave them all
<jamest> ?
<Mr_You> one day we might, not right now, I wish it did it with newest first
<jamest> me too
<jamest> derek: you here?
ajmitch (me@p54-max7.dun.ihug.co.nz) left irc: Ping timeout: 181 seconds
<Mr_You> well congratulations will install 0.1.0 tommorow.. guess I'm going idle
<Mr_You> at some point
<jamest> huh?
<jamest> idle?
<Mr_You> huh?
<Mr_You> yeah
<Mr_You> like I am in here most of the time ;-)
<jcater> hehe
<jcater> added a comment to my own article
<jcater> :D
Action: Mr_You kicks the cable modem.
<jamest> :)
<jamest> and our first bug submission was before we had links posted
<jcater> yip
<jamest> that's scarry
<jcater> scary even
<jcater> wow
<jcater> our site looks much cleaner
<jamest> huh?
<jcater> with the Extra Light theme
<jcater> ooh
<jcater> slash is neat too
<jamest> yeah
<jcater> ooh
<jcater> can I change the default theme to slash ocean???
<jcater> it's blue
<jcater> and clean
<jamest> if anyone gets bored they can update us to nuke 5.3 and kill mysql
<jcater> just needs a logo at the top :)
<jamest> jcater: i don't care
<jcater> mwahahaha
<jamest> i'd update to 5.3 nuke first before I did too much theme work though
<jamest> as themes are IIRC php files
<jamest> and would probably get wiped in an update
<jamest> submitted to freshmeat
reinhard (~rm@62.47.44.7) joined #gnuenterprise.
<jcater> I couldn't resist
<jcater> and change it anyway
<jcater> I'll look at 5.3 nuke later on
<jamest> did you fix the image
<jcater> yes
<jamest> hmmm
<jamest> comes up the same
<jcater> you not like?
<jamest> i dont' see
ajmitch (~me@p39-max4.dun.ihug.co.nz) joined #gnuenterprise.
<jamest> which did you pick?
<jcater> GNUeSlashOcean
<jcater> based on SlashOcean
<jamest> i like it
<jcater> crap
<jcater> it didn't save my change
<jcater> ok
<jcater> there we go
<jcater> it's just less busy
<jcater> when do things show up on freshmeat?
<jcater> the following day?
<jamest> um
<jamest> just later than you post them
<jcater> :)
<jamest> no promises on when 
<jamest> typically by morning
<jcater> anything you need me to do
<jamest> nope
<jamest> i think we're done :)
<jcater> yay
<jcater> nite all
<jamest> nite
jcater (jason@HubA-mcr-24-165-193-222.midsouth.rr.com) left irc: "l8r"
<jamest> nte
jamest (jamest@fh-dialup-201190.flinthills.com) left irc: "[x]chat"
jcater (~jason@HubA-mcr-24-165-193-222.midsouth.rr.com) joined #gnuenterprise.
jcater (jason@HubA-mcr-24-165-193-222.midsouth.rr.com) left irc: Client Quit
reinhard (rm@62.47.44.7) left irc: "Omnis enim res, quae dando non deficit, dum habetur et non datur, nondum habetur, quomodo habenda est.  -- Aurelius Augustinus"
Manaic-yawa (darryl@h24-82-132-209.wp.shawcable.net) left irc: Ping timeout: 181 seconds
ra3vat (~ds@195.239.66.42) joined #gnuenterprise.
<Mr_You> ra3vat!
<Mr_You> 0.1.0! Whooo hooo!
<Mr_You> ;-)
<ra3vat> hi Mr_You
<Mr_You> night
Action: Mr_You &
ra3vat (ds@195.239.66.42) got netsplit.
Mr_You (rouzer@mankind.boredom.org) got netsplit.
dtm (dtm@m206-221.dsl.tsoft.com) got netsplit.
Isomer (dahoose@210-86-56-131.jetstart.xtra.co.nz) got netsplit.
ra3vat (~ds@195.239.66.42) returned to #gnuenterprise.
Mr_You (rouzer@mankind.boredom.org) returned to #gnuenterprise.
dtm (~dtm@m206-221.dsl.tsoft.com) returned to #gnuenterprise.
Isomer (dahoose@210-86-56-131.jetstart.xtra.co.nz) returned to #gnuenterprise.
ra3vat (ds@195.239.66.42) left irc: Ping timeout: 181 seconds
Mr_You (rouzer@mankind.boredom.org) got netsplit.
Isomer (dahoose@210-86-56-131.jetstart.xtra.co.nz) got netsplit.
dtm (dtm@m206-221.dsl.tsoft.com) got netsplit.
Isomer (dahoose@210-86-56-131.jetstart.xtra.co.nz) got lost in the net-split.
dtm (~dtm@m206-221.dsl.tsoft.com) got lost in the net-split.
Mr_You (rouzer@mankind.boredom.org) got lost in the net-split.
ra3vat (~ds@195.239.66.8) joined #gnuenterprise.
Isomer (dahoose@210-86-56-131.jetstart.xtra.co.nz) joined #gnuenterprise.
alexey (alex@techkran.vladimir.ru) left irc: Ping timeout: 181 seconds
alexey (alex@techkran.vladimir.ru) joined #gnuenterprise.
ajmitch (me@p39-max4.dun.ihug.co.nz) left irc: Ping timeout: 181 seconds
ajmitch (~me@p39-max4.dun.ihug.co.nz) joined #gnuenterprise.
nickr (nick@e-172-IP26.empnet.net) left irc: Killed (NickServ (Nickname Enforcement))
ra3vat (ds@195.239.66.8) left irc: "Client Exiting"
ajmitch (me@p39-max4.dun.ihug.co.nz) left irc: "http://www.freedevelopers.net"
Mr_You (rouzer@mankind.boredom.org) joined #gnuenterprise.
dtm (~dtm@m206-221.dsl.tsoft.com) joined #gnuenterprise.
jcater (~jason@w202.z065105010.mem-tn.dsl.cnc.net) joined #gnuenterprise.
jamest (~jamest@hobbes.math.ksu.edu) joined #gnuenterprise.
alexey (alex@techkran.vladimir.ru) left irc: "[x]chat"
neilt (~neilt@user-2ivel3h.dialup.mindspring.com) joined #gnuenterprise.
<mcb30> jamest & jcater: back on #gnuenterprise less than 7 hours after the late-night release?  That's dedication!
<jcater> work calls
<jcater> :(
<jamest> 110 downloads total since the 1am announce
<jamest> 51 of which are windows exes
<jamest> i think taking the time for a setup.exe was well worth it :)
<jcater> yes
<jamest> hey jcater
<jamest> at work already eh?
<jcater> sigh
<jcater> yip
ra3vat (~ds@195.239.66.44) joined #gnuenterprise.
Nick change: mcb30 -> mcb30-lunch
neilt (~neilt@user-2ivel3h.dialup.mindspring.com) left #gnuenterprise.
neilt (~neilt@user-2ivel3h.dialup.mindspring.com) joined #gnuenterprise.
<jcater> I changed the download section at http://www.gnuenterprise.org/download.php
<jcater> it sorts download section be newest files first
<derek> did you release all at same time (me hopes not)
<derek> i.e. announce not release (freshmeat only)
Action: derek thawps jamest and jason
<derek> you guys messed up big time (sigh)
<derek> PLEASE PLEASE PLEASE when making bundled releases
<derek> release all pieces same day, and every BUT freshmeat post all releases
<derek> BUT on freshmeat make the announcements a day a part
<derek> as not doing so cuts number of people who find out about our project by about 30,000 people :(
Action: derek takes blame for this as i forgot to say something about it
<derek> ahhhhhhhhhhhhhh
<derek> what happened to our website?
<derek> the look is way different
<jcater> see what happens when you sleep?
<derek> submitted the release to linux today as a 'press release' kind of thing
<jamest> um
<jamest> freshmeat was olny announce
<jamest> no other sites
<jamest> also, since we're already getting bug reports
<jamest> i expect annother release real soon now :(
ToyMan (~stuq@c5300-2-ip42.albany.thebiz.net) joined #gnuenterprise.
Nick change: mcb30-lunch -> mcb30
ToyMan (~stuq@c5300-2-ip42.albany.thebiz.net) left #gnuenterprise ("Client Exiting").
ToyMan (~stuq@c5300-2-ip42.albany.thebiz.net) joined #gnuenterprise.
ToyMan (~stuq@c5300-2-ip42.albany.thebiz.net) left #gnuenterprise ("Client Exiting").
neilt (neilt@user-2ivel3h.dialup.mindspring.com) left irc: Ping timeout: 181 seconds
ra3vat (ds@195.239.66.44) left irc: Ping timeout: 182 seconds
alexey (~alexey@195.151.214.34) joined #gnuenterprise.
reinhard (~rm@N801P007.adsl.highway.telekom.at) joined #gnuenterprise.
ra3vat (~ds@195.239.66.13) joined #gnuenterprise.
ToyMan (~stuq@c5300-2-ip42.albany.thebiz.net) joined #gnuenterprise.
dneighbo (dneighbo@mail.libertydistribution.com) joined #gnuenterprise.
<mcb30> I've put together a short(ish) document that summarizes what I think is the current status of GNUe (based on various sources and conversations) -
<mcb30> I'd appreciate comments and corrections because there's bound to be some things I've misunderstood -
<mcb30> I've also done some thinking about methods, geas and code-sharing, which is towards the end of the document -
<mcb30> (and I'm sure people will have views on this part! :-) -
<mcb30> You can get it from http://www.fensystems.co.uk/geas-ideas.pdf or http://www.fensystems.co.uk/geas-ideas.tex
Action: mcb30 awaits the flames
<jcater> hmm
<jcater> I get an "Internal Error" with acrobat
<mcb30> try ghostview
<jcater> lemme try another computer to make sure it's not my computer
<mcb30> acrobat often complains about latex-generated pdfs, I find
<dneighbo> sigh
<dneighbo> i have had a document by component for some time
<dneighbo> begging people to keep it updated
<dneighbo> in docbook format
Action: dneighbo is starting to feel unheard
<mcb30> jcater: you can try postscript: http://www.fensystems.co.uk/geas-ideas.ps.gz
<jamest> dneighbo: these are his notes in tryig to understand gnue
<mcb30> dneighbo: I wanted to clear up my understanding - if what I've written is actually worth contributing then I'll merge it into your doc
<dneighbo> did you only have geas-ideas?
<dneighbo> <mcb30> I've put together a short(ish) document that summarizes what I think
<dneighbo>           is the current status of GNUe (based on various sources and
<dneighbo>           conversations) -
<mcb30> the bits that are actually my ideas are mainly methods-related
<dneighbo> is what i was talking about as being duplication :)
<dneighbo> i didnt see that really in your doc
<dneighbo> your doc was geas specific and we have no such document
<dneighbo> well we might but we are highly unorganized on that side (outside of reinhard and neilt)
<dneighbo> that i couldnt tell you if there was duplication there :)
<dneighbo> reading the raw lyx looks like something good to start discussion
<mcb30> it's not completely geas-specific; there's a whole half-page on forms clients!
<mcb30> :-)
<dneighbo> mcb30 well i had to skim tex source so i cant speak to that :)
<dneighbo> trying to get local so i can get in more readable format :)
<mcb30> what format would you prefer?
<mcb30> I can do ASCII?
<jamest>  /msg dneighbo this guy seems sharp, you have any spare shackles laying about we can ship to the UK
<dneighbo> docbook is prefered
<jcater> BUT we take what we can get!
<dneighbo> with output to html/rtf/pdf/ps
<dneighbo> but im NOT complaining
<dneighbo> tex is fine :)
<mcb30> LyX had "issues" with docbook when I tried using it earlier, but I can try again if you actually want to keep this document
<dneighbo> jsut saying i need to bring it to a machine with 'real' tools so i can play :)
<jamest> like a w2k machine
<jamest> that'd rock
<dneighbo> fsck...........
<dneighbo> galeon will not save that file
<dneighbo> it crashed galeon
<dneighbo> you think gv supports remote files?
<dneighbo> like
<dneighbo> gv http://somedomain/somefile.pdf
<jamest> you have wget?
<jamest> wget http://www.fensystems.co.uk/geas-ideas.ps.gz
Action: dneighbo is in process of apt-getting gv and such
Action: dneighbo forgets he did this install 'bare bones'
<dneighbo> no wget (apt is a good friend to have though)
<mcb30> how about lynx?
<dneighbo> ah
<dneighbo> wget + gv seems to have done the trick
<jamest> wget rocks
<dneighbo> man im diong wget for all downloads now
<dneighbo> it is like 1000x faster
<jamest> wget -c http://www.fensystems.co.uk/geas-ideas.ps.gz
<jamest> will continue a interruped download
<jamest> and wget by defaults attempts restart like 10 times
<jamest> think it does ftp too
<dneighbo> argh i cant print from unix wodner if i save as
<dneighbo> if it will fix the issues and make so can open in acrobat on windows
<jcater> grab curl
<jcater> it rocks!
<dneighbo> have a dumb design meeting to go to, when i get back will review
<dneighbo> still didnt specifically see 'forms stuff'
<dneighbo> but regardless this is desgin stuff much more than it is 'status' stuff
<dneighbo> as far as i can tell :)
<mcb30> status of the design! :-)
<dneighbo> which is good :)
<dneighbo> as we need specs for geas and how it will work with forms :)
<dneighbo> and its not duplication of status by product as i thought from your description :)
<dneighbo> http://goats.gnue.org/~dneighbo/gnue/gnue_components/
<dneighbo> was meant as a starter template
<dneighbo> to help 'define' at a high level what the tools are
<dneighbo> and planned on putting some status info within them as well
<dneighbo> with some polish i think this document could be useful as 'the managers introduction to gnue'
<jamest> oxygen is good
<jamest> (old dilbert cartoon's project data sheet after being simplified for upper managment.)
<mcb30> :-)
dtm (dtm@m206-221.dsl.tsoft.com) left irc: Ping timeout: 181 seconds
<dneighbo> dood you MUST read this 
<dneighbo> its so kick butt
<dneighbo> http://biz.yahoo.com/bw/011120/202744_1.html
<dneighbo> basically m$ said they would as punishment for price gouging
<dneighbo> donate money to schools
<dneighbo> like 20% went ot hardware the rest to guess ....
<dneighbo> microsoft software
<dneighbo> redhat steps up and says hell no
<dneighbo> we will give them redhat for free and support for freee
<dneighbo> and donate ALL the money to hardware
<dneighbo> thus giving the schools like 5x more computers with same $ penalty to you :)
<jcater> oh my god
<jcater> that rocks
<dneighbo> now if that aint playing like m$ does :)
<jcater> it'll never go
<dneighbo> and thats REAL punishment
<jcater> but that'd be awesome
<dneighbo> jcater: i know but its GREAT to see this on yahoo
<dneighbo> as ANY american reading this
<dneighbo> couldnt be AGAINST it
<jcater> :
<dneighbo> i.e. against schools getting more for the money
<jcater> :)
<jamest> lol
<jcater> hmm
<jamest> rofl
<jcater> time to get that education management module written
<dneighbo> this is just classic
<dneighbo> ``While we applaud Microsoft for raising the idea of helping poorer schools as part of the penalty phase of their conviction for monopolistic practices, we do not think that the remedy should be a mechanism by which Microsoft can further extend its monopoly,'' said Matthew Szulik, CEO of Red Hat. ``Through this proposal all of the states and all of the schools can win, and Microsoft will achieve even greater success for its stated goa
<dneighbo> totally m$ tactics :)
<dneighbo> a. compliment the original person
<dneighbo> b. make an off handed statement that is a total slam with a smile
<dneighbo> c. and then shove the stake through the heart
<dneighbo> difference is RH is using the truth to do it
<dneighbo> m$ usually uses lies :)
alexey (alexey@195.151.214.34) left irc: "Client Exiting"
<jamest> well
<jamest> it's really funny in that the total cost to distribute that software is media only for both companies
<dneighbo> i love the ...and m$ will achieve even greater success for its 'stated goal' of helping schools
<dneighbo> as its calling m$ on the carpt
<dneighbo> carpet
<jamest> and at MS's volume probably means they would pay less for the media
<dneighbo> meaning they make is sound like they are doing this big favor to help the schools like they REALLY care about the schools
<jamest> so it really puts the whole value of the package on the table
<dneighbo> now red hat has called them on it, if you really care about the schools give them more
<dneighbo> if you really care about just deferring the payment to you, give them your software :)
<reinhard> mcb30: current state we see geas not providing lookup fields directly
<reinhard> so in your example you would get the country name by getting the object reference and then loading the country name from that referenced country instance
<mcb30> ok, but are we still wanting to provide lookup fields?
<mcb30> i.e. is it a "not yet implemented" or a "discarded" feature?
<reinhard> from my pov no
<reinhard> discarded imo
<reinhard> i would rather say we could want to provide some function like
<reinhard> if the class "customer" has a field "country"
<reinhard> which is a object reference to the country class
<reinhard> and the country class has a field "text"
<reinhard> then we could want to access that field as
<reinhard> "customer.country.text"
<mcb30> how would you assign to customer.country under that scheme?
<reinhard> setField ("customer.country", my_object_reference)
<reinhard> and
<reinhard> setField ("customer.country.text", something)
<reinhard> would issue an exception like "cannot assign to that field"
<reinhard> but you could do getField ("customer.country.text")
<mcb30> so does that mean the object reference fields are meant to be read-write?
<mcb30> I thought they were read-only
<reinhard> my understanding is they should be read-write
<reinhard> because my understanding is also the id fields are actually hidden
<reinhard> so you can do
<reinhard> my_country_object = (get the damn object somehow);
<reinhard> my_customer.setField ("country", my_country_object);
<reinhard> which would in fact get the id value of my_country_object
<reinhard> and set the hidden field customer._country to that id in the database
<mcb30> ok, that all looks good, but
<mcb30> how do you use that through a non-OO interface (i.e. suppose we create an ODBC driver to GEAS)
<mcb30> ?
<reinhard> i guess we just don't want that
<reinhard> at least that's true for me :)
<reinhard> mho is that either we want an object oriented interface
<reinhard> or a relational interface
<reinhard> but we don't want both
<mcb30> why not both?
<reinhard> <disclaimer>this is my personal understanding of things</disclaimer>
<reinhard> it is a contradiction imo
<reinhard> isn't it?
<mcb30> I don't know - I'm thinking about it now!  :-)
<mcb30> it's just that some parts of the object approach maps quite well onto the relational approach (e.g. classes<->tables, instances<->rows) - I'm interested to see if we can extend that all the way
<mcb30> still thinking...
<reinhard> the main difference between relational and oo are not the objects/tables for themselves
<reinhard> but the way they are linked together
<mcb30> that's the bit I'm thinking about... :)
<mcb30> thought: we are planning on using relational DBs to store the objects anyway, so therefore there exists a relational way to link the data together (otherwise the whole project is impossible!)...
<reinhard> yes
<mcb30> therefore it should be a case of simply exposing the relational links in a friendly way....?
<mcb30> at the same time as providing the OO mapping...?
<mcb30> am I talking garbage here?
<reinhard> yeah
<reinhard> um
<reinhard> yeah to the first 2 lines not to the last one :)
<mcb30> :)
<reinhard> but that would be 2 different interfaces actually
<mcb30> yes
<reinhard> ok
<reinhard> i think now we are on the same page
<reinhard> we can of course have 2 different interfaces
<reinhard> one providing a relational view and the other the oo view
<reinhard> but for the relational view i'm not sure for what we need geas anyway
<reinhard> and why not accessing the db directly for that
<mcb30> jamest / jcater: how does the forms DB interface work - does it expose the data as object-oriented or relational?
<mcb30> half-way down page 3: "The role of GEAS"
<reinhard> another point i see
<mcb30> I was thinking of the code that provides the data interface being part of gnue-common rather than just GEAS
<mcb30> (i.e. the code we've just been talking about)
<reinhard> i think we _don't_ want to use stored procedures
<reinhard> even if the database would support them
<mcb30> what's the reasoning?
<reinhard> i think you write methods either in python or in sql
<reinhard> isn't that true?
<reinhard> and neither seems portable to the other world
<mcb30> have you read the part where I suggest creating a "methods language"?
Action: mcb30 awaits lots of flames now
Action: reinhard is still on page 3
Action: reinhard is reading that section about methods language
Action: reinhard shudders
Action: mcb30 cowers
Action: reinhard takes a deep breath
<reinhard> ok
Action: reinhard takes another deep breath
<reinhard> i definitively like the idea of a single method language that is translated into whatever the method executer understands
<reinhard> i like it
<reinhard> yes
Action: mcb30 is relieved
<reinhard> as long as i don't have to implement that :)
<mcb30> :-)
<reinhard> i'm not sure what i should think about this trend
<mcb30> I kind of assumed that by suggesting it I would implicitly be volunteering to code it
<reinhard> that 2tier and ntier seems to grow togehter slowly
<reinhard> 2 tier and n tier were regarded as two complete different worlds once
<reinhard> and the distance between them seems to fade away
<reinhard> i just don't know which one of the two is moving towards the other :)
<mcb30> both moving would probably be good
<reinhard> exactly this is what i'm not sure about
<reinhard> 2-tier moving towards n-tier means 2-tier looses its simplicissity
<reinhard> n-tier moving towards 2-tier means n-tier looses its power and scaleability
<reinhard> (probably)
<reinhard> (maybe)
<reinhard> sorta
<reinhard> kinda
<reinhard> dunno
<reinhard> arrrgggh
<mcb30> I don't think that should happen, but it would be good to share code where possible
<reinhard> yeah
<reinhard> sure
<reinhard> where possible and reasonable
<mcb30> of course
<reinhard> but i don't think 2-tier should be identical to n-tier just that all code runs on same machine
<mcb30> I'm not actually suggesting huge changes to 2-tier - check point 4 on page 5
<mcb30> basically the change would just be to the language that the triggers were written in
<reinhard> i'm really sorry
<mcb30> sorry?
<reinhard> i think i am starting to get an emotional tie to geas
<reinhard> coming from the amount of time i spent with it
<reinhard> and i get the feeling that i loose the ability
<reinhard> to think 100% rationally about such issues
<reinhard> i definitively like the idea of that method language
<reinhard> i am sure about that
<reinhard> what i don't like is the overall feeling (not just now but continuing for several weeks/months)
<reinhard> that geas is in the process of loosing importance continously
<reinhard> which is a 100% irrational thing from my side
<jamest> huh?
<jamest> not at all
<jamest> geas is the cornerstone of the n-tier gnue
<reinhard> i'm just like a father that wants his son to be the best pupil in class :)
<jamest> we just want to gut it lots of it
<jamest> to put in a common lib
<reinhard> jamest: i know that
<jamest> a tip - do what i do
<reinhard> really
<mcb30> There is lots of important stuff in geas that should probably be in common so that everything else can benefit from it
<jamest> when jcater gutted gnuef i took secret pleasure in watching gnuef distro size shrink
<jamest> it sick how small designer is
<jamest> i just thought of something
<jcater> what is sick is that I based designer on forms
<jcater> :)p
<reinhard> :)
<jamest> (yeah that's real sick)
<jamest> while some people boast 750,000 lines of code
<jamest> here I am hoping to make our code base smaller
<jamest> i must have something wrong with me
<jamest> er, something _else_ wrong with me
<reinhard> lol
<jcater> well, I've always heard that a sign that a product is maturing is that it's code base starts shrinking instead of growing
<jcater> of course, there's no way forms is mature :)
<reinhard> yeah that's an old chinese proverb iirc
<reinhard> something is perfect not when there's nothing more to add
<reinhard> but when there's nothing more to take away.
<jcater> ah
<jcater> yes
<jcater> that's where I heard it from
<reinhard> funny what the old chinese knew about software :)
<jcater> yes
<jamest> 209 downloads
<jamest> no bug reports
<jamest> hmmm, they must not be able to install it
<jamest> 78 win32 downloads
<jamest> ~48000 hits on the site
<jamest> "It's alive!"
<jamest> l8r everyone
jamest (~jamest@hobbes.math.ksu.edu) left #gnuenterprise.
ra3vat (ds@195.239.66.13) left irc: "Client Exiting"
od (~del@host.182.75.23.62.rev.coltfrance.com) joined #gnuenterprise.
od (del@host.182.75.23.62.rev.coltfrance.com) left irc: Client Quit
<dneighbo> i am going to be devils advocate and say
<dneighbo> yes lots of stuff from geas should be in common
<dneighbo> which means lots of geas code goes away
<dneighbo> in which case short of gcd's geas almost becomes a python application
<dneighbo> just noting that
<dneighbo> and awaiting flames
<dneighbo> anyone here?
<dneighbo> mcb30 you state: a method will correspond to a stored procedure in the database
<dneighbo> i dont think i agree with that
<dneighbo> depending on how you mean 
<dneighbo> if you mean in the two tier world its like calling a stored procedure in a db (in concept) i agree
<dneighbo> if you mean that the method is actually stored as a stored procedure in the db i strongly disagree
<dneighbo> as what i see you saying here, is that one would write something in a GNUe defined language and have it converted into a the SQL DB Stored Proc langugage
<dneighbo> which i think defeats MANY of the goals of geas
<dneighbo> but maybe i am misunderstanding
<dneighbo> reading your document further....
reinhard (rm@N801P007.adsl.highway.telekom.at) left irc: Ping timeout: 181 seconds
<dneighbo> sounds like you are actually wanting to split 
<dneighbo> so you can write methods as db specific triggers (via a new ddl)
<dneighbo> or in geas as its understood now
reinhard (~rm@62.47.44.7) joined #gnuenterprise.
<dneighbo> which might not be an all bad idea, though the maintenance of said implementation makes me cringe
<dneighbo> as too bison flex doing the work
<dneighbo> the MORE i think about things i think we shoudl be using xml
<mcb30> am back, reading your comments...
<dneighbo> with xslt its a no brainer to make it look human readable
<dneighbo> and makes conversions into other formats VERY easy
<jcater> what in xml?
<dneighbo> as well as making self documenting via dia or other toos a cinch
<dneighbo> gcd's
<dneighbo> gnue class definitions
<jcater> ooh
<mcb30> how easy is it to write "code" in XML?
<dneighbo> originally i was against it because they are more like code
<jcater> that would also mean Designer could support them relatively easy
<dneighbo> and coding in xml is ugly
<dneighbo> but after playing with xslt
<dneighbo> i think what we are losing istn worth it
<dneighbo> and making them xml would be much better in the long run
<dneighbo> as they would still be editable by hand
<dneighbo> but they would be come self documenting
<dneighbo> and if you wanted to create say corba objects or other things off them etc
<dneighbo> its as easy as writing a stylesheet
<dneighbo> you could even write a style sheet to out put them to their current format
<dneighbo> anyhow just talking out loud
Action: mdean likes what he reads...
<dneighbo> made many other comments
<dneighbo> hopefully people respond
<mcb30> how would you write something like "sum = a + b" in XML?
<dneighbo> depends on how you want to do it
<dneighbo> <function type="sum>
<dneighbo>   <value>a</value>
<dneighbo>   <value>b</value>
<dneighbo> </function>
<dneighbo> could be one way
<dneighbo> or
<dneighbo> <sum>
<dneighbo>   <value>a</value>
<dneighbo>   <value>b</value>
<dneighbo> </sum>
<dneighbo> etc etc etc
<mcb30> now, to me, "sum = a + b" looks much nicer
<dneighbo> mcb30 yes it does
<dneighbo> but if you do it xml
<dneighbo> in aobut 30 seconds
<dneighbo> you can make it valid code
<dneighbo> for python, perl,php, javascript etc
<dneighbo> as well as make html docs
<dneighbo> postscript docs
<dneighbo> etc etc etc
<dneighbo> and make a dia diagram 
<dneighbo> i.e. its a lot more flexible
<dneighbo> and you could make something that displays it as
<dneighbo> sum = a + b
<mcb30> but you could produce the XML from "sum = a + b" anyway, and
<dneighbo> also gnue is designed for end users not hackers
<dneighbo> and imagine 98% of people using it will use an IDE
<mcb30> it would be much nicer from the editing point of view to be able to edit sum=a+b rather than a four-line XML structure
<dneighbo> and so they will simply be typing
<dneighbo> sum=a+b and all tags are handled for them
<dneighbo> i.e. i agreed with you a month ago
<dneighbo> now i dont
<dneighbo> part of this is i see how quickly designer matured
<dneighbo> so i know i wont have to edit them in emacs for life
<mcb30> I don't like the idea of deliberately making it difficult for a hacker to write methods
<dneighbo> and part of it is i see the power of xml transformations
<dneighbo> and see a lot of work
<dneighbo> mcb30 : my personal opinion is 80 / 20 rule
<dneighbo> 80% of gnue users will not be 'hackers'
<mcb30> 80% of gnue uses will probably not be the ones writing the methods
<dneighbo> i disagree
<dneighbo> i mean 80% of gnue CODERS will not be hackers
<mcb30> hmmmm - has anyone played with the Morphic environment within Squeak smalltalk?
<dneighbo> smalltalk is not an option within gnue :)
<mcb30> If we were planning to implement something like that then I would concede on storing methods as XML data
<dneighbo> whoa there charlie
<mcb30> I know we won't use smalltalk - it's just an example of a perfect drag-and-drop coding interface
<dneighbo> two distinct issues
<dneighbo> im talking gcd's
<dneighbo> not method code
<dneighbo> last time i looked at geas
<mcb30> I was talking method code, that's why I said "sum = a + b"
<dneighbo> gcd's didnt contain the code
<dneighbo> it contained a pointer to the code file
<dneighbo> i woudl say gcd's are prime for xml
<dneighbo> method codes would be the native language
<mcb30> ok, no problem there
<dneighbo> UNLESS you went down the idea of ddl for stored procs
<dneighbo> i would say THAT kind of method and only that kind would be XML
<mdean> they would translate nicely to wsdl as well
<dneighbo> we have ddl already in xml
<mcb30> we do?
<dneighbo> that does database creates etc
<mcb30> where did that come from?
<dneighbo> that supports lots of db's
<dneighbo> mdean and my cattle prod
<mcb30> isn't that a major overlap with GCDs?
<dneighbo> it would just be a matter of extending it to add trigger/stored proc support
<dneighbo> mcb30 well in two tier gcd's dont exist
<dneighbo> in current state
<dneighbo> and it was done for DCL (another project)
<dneighbo> but yes thats the point
<mcb30> no, but the functionality overlaps quite a lot
<dneighbo> right now geas is hardcoding all that crap
<dneighbo> we want it to use the ddl
<dneighbo> the idea is to include the ddl in common
<dneighbo> and get geas to use the common-db system
<mcb30> is there an example ddl file in cvs I can look at?
<dneighbo> and then it has its problems solved :)
<mdean> mcb30: http://cvs.sourceforge.net/cgi-bin/viewcvs.cgi/dcl/dclgw/scripts/install/xml/
<dneighbo> http://cvs.sourceforge.net/cgi-bin/viewcvs.cgi/dcl/dclgw/scripts/install/xml/
<dneighbo> or what he said :)
<dneighbo> they will be in gnue cvs soon as well
<dneighbo> was on my todo this weekend :)
<dneighbo> of course only the .py files the .psp files
<dneighbo> and the style sheets
<mcb30> looking now...
<mcb30> ok: does this approach provide all of the functionality of the parts of GEAS that interpret the GCDs and create the database tables?
<mcb30> can't easily tell how complete it is from a quick CVS browse
<dneighbo> um mcb30 not familiar enough to talk to that
<dneighbo> it should be able to replace the CREATE parts of geas
<dneighbo> i.e. geas should be able to call these to create the db's
<dneighbo> in a since
<dneighbo> sense
<dneighbo> what i see happening is in geas code
<mcb30> reinhard: are you still here and what do you think of all this?
<dneighbo> it takes the requirements and builds an XML tree in memory
<dneighbo> then based on the users db choice
<dneighbo> uses the .xsl file to create the db creation script (which could be done in memory)
<dneighbo> and executed out
<dneighbo> i kind of like the idea of it writing the .sql file (as an audit trail)
<dneighbo> then loading the .sql files
<dneighbo> but need to look
<dneighbo> the other option with geas is to use common-db
<dneighbo> and make common-db support fully introspection to dbs and all its supporting thigns to create tables and such
<dneighbo> and then call the unified api and let common-db sort out the implementation
<dneighbo> thus nulling the ddl work
dtm (~dtm@m206-221.dsl.tsoft.com) joined #gnuenterprise.
<dneighbo> but anyhow just thoughts
<mdean> ja - especially if introspection matches the XML DDL format and/or types
<dneighbo> ignore my rants as i appear to be grossly out of touch
<dneighbo> lots of communication goes on outside this channel or this channel when im not around and have been too sloppy to keep up
<dneighbo> mdean you still vacationing?
<dneighbo> er holidaying
<mdean> all week :)
nickr (nick@e-172-IP26.empnet.net) joined #gnuenterprise.
<mcb30> Just MHO, but I think it would be good to try to reuse code for DB creation from GEAS, even if this means translating it to python and making it work with an XML data structure instead of a flex/bison derived structure
<mcb30> after all, that code in GEAS works
<mcb30> already
<mcb30>  (as far as I can tell - it worked for me :-)
<reinhard> mcb30: i am still here
<reinhard> um to be precise s/still/again/ :)
<reinhard> mcb30: the code in GEAS _does_ work
<reinhard> .gcd files are not xml
<reinhard> and we had this discussion about whether they should be xml or not
<reinhard> we had the discussion 3 or 4 times already
<reinhard> .gcd files are not xml
<dneighbo> yes i stated that :)
<dneighbo> that a. they werent and b. we had the discussion
<reinhard> and they will not be as long as they are authored by hand
<reinhard> and the idea was that as soon as they are not any longer authored by hand
<reinhard> there will be no more need for .gcd files
<reinhard> but the information will rather be put into a database
<reinhard> and we will use xml to get the data out of database a and into database b
<reinhard> xml will be the interface between 2 systems
<reinhard> but not between a system and a user
<reinhard> .
<nickr> damnit, docbook doesn't work
<mcb30> on the methods side, I'm still confused over whether or not people think method *code* should be written in XML...
<dneighbo> depends on what kind of method code
<dneighbo> if you are talking the intermediate server method code (From your paper) i woudl say no, it should be in whatever language you are using, python, perl etc
<dneighbo> if you are talkign the single source code that gets turned into db triggers/stored procs then i would say yes
<dneighbo> im still not thoroughly conviced i like the idea of doing the latter at all xml or otherwise, but we did i think doing in xml could be advantageous
<mcb30> What are triggers coded in atm - is it python?
<dneighbo> i am but only one of many with an opinion :)
<dneighbo> python or C are supported i think for 'methods'
<dneighbo> we do NOT support db native triggers/procs
<mcb30> I'm talking specifically about the triggers in forms here
<mcb30> are they python?
<dneighbo> python only for forms triggers
<dneighbo> 'currently' plans are in works to change that
<mcb30> so the designer support for triggers is a case of allowing you to write some python code, yes?
<dneighbo> yes
<mcb30> just checking
<dneighbo> in loose terms
<dneighbo> in designer on a widget
<dneighbo> you can choose (an event) for lack of better word
<dneighbo> pre-focusin prefocusout etc
<dneighbo> and then tie a trigger to to that
<dneighbo> so taht when that even happens that trigger is used
<dneighbo> you can do a 'shared trigger'
<dneighbo> which lets you make the trigger more like a function/component
<mcb30> ok, skipping back to generic method languages and the XML issue:
<dneighbo> so that you can tie it to more than one event by name
<mcb30> explain to me the advantages of using XML for this
<mcb30> compared to using flex/bison
<dneighbo> um instead of a bunch of flex bision code you end up with a nice clean .xsl file for each db implementation
<dneighbo> and it serves better as an INDEPENDENT tool
<dneighbo> part of the goal of GNUe is all tools work together or separately
<dneighbo> we have ddl that does database schemas
<dneighbo> if we made it support database storedprocs/triggers
<dneighbo> we make the ddl tool better while solving an issue within our appserver at same time
<dneighbo> which means in theory other projects will use it that may not use our server
<dneighbo> which means more users/testers/coders which means a better product
<dneighbo> if im making any sense i have no idea
<mcb30> what about the dependency issue
<dneighbo> i dont hear anyone supporting me so i suppose im not
<dneighbo> what dependency issue?
<mcb30> I don't think anyone else is listening atm
<mcb30> what code do you need in order to apply XSL?
<reinhard> i am not listening, but i am watching :)
<dneighbo> pretty much xml is a requirement to use gnue
<dneighbo> an xslt parser might be an additional dependency
<dneighbo> but one that will be necessary for reports
<dneighbo> and eventually it is rather possibly that it will be included as part of pyXML (if it isnt already)
<dneighbo> which we already have as a dependency
<mcb30> ok
<dneighbo> im not trying to trivialize that and im not saying we dont care if we pile up dependencies
<dneighbo> i think its a decent point
<dneighbo> BUT remember if gcds were made xml then you lose a flex/bison dependency
<dneighbo> and you have an even swap ;)
<reinhard> no
<mcb30> flex/bison is not required when *using* the gcds
<mcb30> you don't even need them to compile the code
<reinhard> not even required for compiling
<dneighbo> mcb30 ? what then parses the gcd?
<reinhard> yes what he said :)
<mcb30> you just need them to generate the code in the first place
<mcb30> i.e. if you do a "make maintainer-clean"
<mcb30> then you will need them
<mcb30> otherwise you won't
<reinhard> of if you compile from cvs
<dneighbo> what parses the gcd?
<reinhard> a c program
<mcb30> flex/bison take the grammar definition...
<reinhard> that is generated from the flex/bison file
<mcb30> and create a program capable of parsing that grammar
<reinhard> but it's normal c code
<mcb30> what reinhard said!
<dneighbo> right, now you are talking semantics though
<mcb30> no
<dneighbo> if flex/bison didnt generate the grammar
<mcb30> it's like saying that autoconf and automake are a dependency
<dneighbo> then the c program wouldnt work right?
<mcb30> you never need autoconf/automake to build a project
<mcb30> you only need them if you want to regenerate the configure scripts
<mcb30> it's the same
<dneighbo> well currently this is how the ddl works
<dneighbo> you only need xslt if you are a developer
<dneighbo> as you would do the transformation and then distribute the output
<dneighbo> BUT 
<dneighbo> it woudl be different in the geas case of doing things
<reinhard> a developer of what=
<reinhard> ?
<dneighbo> nevermind
<mcb30> I'm not arguing against using xslt on dependency grounds - I just wanted to point out that there are no dependency issues with flex/bison
<mcb30> either
<mcb30> ok, slight tangent:
<mcb30> assume for a minute that designer is 100% complete...
<mcb30> and that I want to use it to create a class...
Action: dneighbo is waiting for jason to laugh
<jcater> :)
<mcb30> and that this class contains the omnipresent "sum=a+b" field...
<mcb30> what does the interface look like that enables me to enter this calculated field?
<mcb30> i.e. can I type "sum=a+b" somewhere?
<mcb30> or do I build it up using graphical blocks, or what?
<jcater> mcb30: however we want to do it
<jcater> that'd have to be discussed, but designer could do either
<jcater> or both
<mcb30> I'm thinking of how it would generate the method definition if the method definition is in XML
<mcb30> how would it do this from a statement such as "=a+b"
<mcb30> wouldn't this require something like a flex/bison parser anyway?
Action: jcater is not really here.. just saw designer in the logs
<dneighbo> sigh i dont think so
<dneighbo> as to designer
<dneighbo> i FULLY expect designer to really be the GNUe IDE at somepoint
<dneighbo> where you would write your methods code etc
<dneighbo> as to plugging it into forms
<dneighbo> i expect it to have object introspection to geas
<dneighbo> so it doesnt give a rats ass
<dneighbo> i.e. you are binding the method
<dneighbo> then if you wish to look at the code to support that method you would probably be given either the db specific code
<dneighbo> or the xml code or have a choice
<dneighbo> i hate to dwell too much on this
<dneighbo> because until now we havent really discussed doing db specific stuff
<dneighbo> which moots the whole issue of xml methods altogether if we decide to not support it
ra3vat (~ds@195.239.64.51) joined #gnuenterprise.
<dneighbo> i think it makes gnue more complex than it needs to be
<dneighbo> and makes distribution and maintenance more difficult
<mcb30> in what way?
<dneighbo> though it would provide some performance gains
<mcb30> it would provide reliability gains,
<mcb30> it would provide flexibility gains,
<dneighbo> i dont see how it makes for a reliablity gain?
<mcb30> it would cut down the amount of method code that needed to be produced,
<dneighbo> or a flexibility gain (well this is half and half)
<dneighbo> how do you figure?
<dneighbo> either you have to do something or you dont
<dneighbo> how does it cut down on the code?
<mcb30> reliability: you are guaranteed that a method executing in a forms client functions exactly the same as equivalent code running in whichever backend you choose
<dneighbo> sigh
<dneighbo> that is irrelevant
<mcb30> why irrelevant?
<dneighbo> sigh
<dneighbo> the idea of an app server is to not use db specific trigger/proc
<dneighbo> but rather make objects/components instead
<mcb30> "backend" is used here to include geas
<reinhard> my understanding is that mcb30 is talking about
<dneighbo> i dont think we are speaking on the same thing
<mcb30> quite possible not
<mcb30> hold on,
<reinhard> sharing the same method definitions for 2 tier and 3 tier
<dneighbo> lets talk with fact that database stored procs and triggers do not exist at all
<dneighbo> reinhard sigh...
<reinhard> mcb30: is my understanding correct?
<dneighbo> i give up... im not able to communicate in this medium currently i guess
<reinhard> or close to that?
<mcb30> dneighbo: I'm talking about the contents of section 4 within the document I posted...
<mcb30> ...you can omit the bit about methods as stored procedures in e.g. PostgreSQL if you want
<dneighbo> the document im reading talks about using database specific triggers/stored procs and abstracting it by creating a home grown language 
<mcb30> but the rest still holds
<dneighbo> i dont have document readily available now
<mcb30> using db specific triggers / stored procs is one option...
<dneighbo> so dont know section 4
<mcb30> ok - wait a sec
<dneighbo> but what it comes down to is what ever gets commit will get used
<mcb30> reinhard: yes, that's what I mean
<dneighbo> and no one else has an opinion or is not here
<reinhard> thanks :)
<dneighbo> so whatever you all wish to implement feel free
<dneighbo> btw: this reminds me of original geas discussions with andrew
psu (psu@manorcon.demon.co.uk) joined #gnuenterprise.
<mcb30> want to give me some history?
<dneighbo> and people then curse me later that i refuse to recognize geas
<dneighbo> and i dont use it in samples
<dneighbo> and i dont promote it to people
<reinhard> ok
<dneighbo> and when i say its not useful to me thats why
<dneighbo> people get mad
<reinhard> can you give in a nutshell how you see the role of geas?
<dneighbo> reinhard apparently not :)
<dneighbo> i have tried but dont communicate well i guess
<dneighbo> i need a magic startrek 'translator' device :)
<mcb30> :-)
<psu> I think you do yourself down Derek
<psu> one of the problems I have with KC 
<psu> is trying not to over-quote you
<psu> you give good copy ;-)
<dneighbo> now also note im not saying im correct
<dneighbo> im just trying to state how i see my needs
<dneighbo> i think most business folks are probably in a similar need situation as me (though i could be delusional)
<dneighbo> unfortunately i wish jcater and/or jamest were able to partake in discussion as well
<dneighbo> fwiw: i think this is good conversation
<dneighbo> what geas IS
<mcb30> jcater: wake up!
<dneighbo> really needs to be discussed 
<dneighbo> :)
<mcb30> designer
[14:35] Last message repeated 3 time(s).
<dneighbo> regardless of what language implementation, dependencies etc
<mcb30> maybe he'll notice that - worked last time :-)
<dneighbo> it has from the beginning never been well defined even on a high level
<dneighbo> what will it do
<dneighbo> jcater: designer 
[14:36] Last message repeated 5 time(s).
<dneighbo> thats not a knock on andrew etc either
<dneighbo> they had some goals for their project we rode on the coattails
<dneighbo> i think they introduced some concepts we didnt see as part of the scope
<dneighbo> and we have discovered somethings we left out of our original thoughts etc
<dneighbo> in a NUTSHELL what i see GEAS being
<dneighbo> middleware that operates independently of the client or the datastorage
<dneighbo> specifically
<dneighbo> it allows for remote communications to both the client and the datastorage
<dneighbo> it should allow for remote method invocation
<reinhard> brb
<dneighbo> i see its key functionality as two primary thigns
<dneighbo> a. providing the client ONE api to deal with data and datastorage (this can be via objects or not)(storage abstarction can be part of geas or not)
<dneighbo> b. provide way for client to store business rules for maximum reuse and portability (i.e. its not part of the storage(db) so its not 'vendor' specific)
<dneighbo> part of b is getting rid of a vendor dependency (the norm for db triggers/procs)
<mcb30> that matches almost exactly what I wrote in the doc:
<dneighbo> the other part is making it so integrator, reports, etc can all use the same rules
<dneighbo> so if you say 'is account valid'
<dneighbo> EVERYTHING uses the same logic :)
<mcb30> Two primary functions:
<mcb30> * Aggregation of database backends. GEAS would provide a single "virtual database" in which the actual data might be spread over several different database backends (PostgreSQL, CVS, etc.).
<dneighbo> well that is highest level (my rant above)
<dneighbo> teh second level items were
<mcb30> * Support for methods (i.e. business rules)
<dneighbo> -- wanted to be usable by programmers so that methods are what they like... (so we support pythno, perl, etc etc) and we would NOT create our own custom language
<dneighbo> -- that the rpc method woudl be flexible to let developer choose transport
<mcb30> I'm only suggesting use of a custom language where it makes sense:
<mcb30> i.e. calculated fields using simple formulae are a prime example,
neilt (~neilt@user-2ivem6b.dialup.mindspring.com) joined #gnuenterprise.
<mcb30> because then you can write the code once and have it available in any time of front end (wxPython, browser etc.),
<mcb30> s/time/type/
<mcb30> and the same code is used by GEAS to calculate the field
<mcb30> but something more complicated
<mcb30> (e.g. lookups of addresses from postcodes etc.)
<mcb30> - the sort of thing that you would not expect to be able to execute inside a client without talking to GEAS -
<mcb30> would only ever be executed inside GEAS, and so
<mcb30> could be written in any language of the programmer's choice
<mcb30> since portability between GEAS and the clients would not be an issue
<mcb30> anyone still here?
<reinhard> just a sec
Action: psu is - KC-ing in real time
<dneighbo> mcb30 well the issue was (we had this discussion in forms triggers)
<dneighbo> was namespacing or a form of short cutting
<dneighbo> for common simple things
<dneighbo> like sums etc
<dneighbo> so this really to me is a 'trigger language' implementation issue
<dneighbo> there is a file that discusses this somewhere
<dneighbo> my big hang up was making methods storedprocs/triggers in the db
<dneighbo> btw: im not fully against that
<dneighbo> just not something i envisioned up front
<mcb30> i fully understand why you would not want to go vendor-specific and have storedprocs/triggers written in the db native language
<dneighbo> i dont see tons of value but jcater and i had a similar discussion one time
<dneighbo> and there maybe value in have this 'option'
<dneighbo> i.e. it woudl not be a RECOMMENDED way to do thigns
<dneighbo> but if you wanted to squeeze performance
<dneighbo> it might be worth it to make some that way 
<mcb30> yes, performance is probably the biggest potential beneficiary
<dneighbo> basically im torn on it
<dneighbo> if i had my choice i would say ignore it for initial implementations
<dneighbo> for simplicity sake
<dneighbo> but leave the door open to add it later
<mcb30> the beauty of having a methods language that can be translated into JavaScript, python, PL/PGSQL etc is that you lose the vendor-specific trap
<dneighbo> yes but then you have done the cardinal sin
<mcb30> which one? :-)
<dneighbo> you have made someone learn your language :)
<dneighbo> most hackers dont mind changing from one language to another etc
<mcb30> not necessarily if you can design them using a drag-and-drop interface in designer :-)
<dneighbo> but walk into MOST IT shops
<jcater> mcb30: the downside is you lose the advantages of each of those languages as well
<dneighbo> and they would lynch you if you asked them to use something other than what they use now :)
neilt (neilt@user-2ivem6b.dialup.mindspring.com) left irc: Ping timeout: 181 seconds
<dneighbo> mcb30 : btw: we debated this as part of trigger spec
<mcb30> I think I know the document
<mdean> dneighbo: have you ever used DTS in MSSQL?
<dneighbo> no
<mdean> do you have 7 or greater?
<dneighbo> have heard of it, but didnt use mssql long enough
<mdean> ah
<dneighbo> i stopped REAL work on mssql at 6.5
<mcb30> is it gnue-common/doc/TriggerSpecifications.txt?
<dneighbo> and only 'toyed' with mssql 7.x
<mdean> yeeesh - 6.x sucked
<dneighbo> mcb30 yes
<dneighbo> mdean how dare you say that about sybase :)
<mdean> I envision a designer for this type of work to be somewhat like DTS
<dneighbo> i envision gnue designer :)
<mdean> 6.x is M$ only - no Sybase
<dneighbo> mdean 6.x is pretty much sysbase
<dneighbo> having used both during that time
<mdean> no it isn't - you're talking SQL 4.x and back for that
<dneighbo> different face pretty darn similar engine :)
<mcb30> can you describe DTS for us poor coders who haven't ever seen or heard of it?
<dneighbo> mdean even 6.x was pretty similar, it wasnt a clone like 4.x but it was close
<mdean> DTS stands for Data Transformation Services
<dneighbo> now 7.x is 100% different animal :)
ToyMan (~stuq@c5300-2-ip42.albany.thebiz.net) left #gnuenterprise ("Client Exiting").
<mdean> it allows you to drag and drop objects in a workspace
<mdean> objects being almost anything
<mdean> server connections
<mdean> database connections
<mdean> activex scripting
<mdean> email support
<mdean> and it all has workflow built in
<mdean> by building conditions between the objects
<mdean> flow can also be halted or continuued based on your preference
<mcb30> any similar products I might have come across? (any free ones?)
<mdean> no
<mcb30> damn
<mdean> M$' big claim on this is that Oracle shops will by MSSQL 7 just to have DTS
<mdean> DTS is really nice - lemme see if I can dig up a link that has some pics
<mdean> I've made a few nontrivial packages with it
<mcb30> I was just about to ask for a screenshot or two :-)
<mdean> dneighbo: this would also be applicable to Integrator - since it is the primary use of DTS
<dneighbo> yes
<dneighbo> i see integrator as basically geas + reporter
<dneighbo> with a nice ui :)
<dneighbo> source -> reporter -> mapping ui -> reporter -> geas --> source
<dneighbo> or something 
<dneighbo> anyhow dnot want to go there right now :)
<dneighbo> the big unit here just bought versata? or something
<dneighbo> and they are all creaming their jeans
Action: dneighbo is curious to see it
<dneighbo> its all java based
psu (psu@manorcon.demon.co.uk) left #gnuenterprise.
<dneighbo> but gives idea what CORPORATE IT folks are looking at in large interprise
<dneighbo> i know its EXPENSIVE as all get out
<mdean> I'll fire up the laptop and take my own screenshot ;-)
<dneighbo> rofl
<dneighbo> sure thats 'legal'
<Mr_You> hi
<dneighbo> the EULA might prevent that
Action: dneighbo wishes he were kidding
<Mr_You> What's new in GNU Bayonne: http://www.newsforge.com/article.pl?sid=01/11/21/1945250&mode=nocomment
<mdean> dneighbo: not like I'm publishing benchmark results
<dneighbo> http://www.versata.com/
<dneighbo> mdean do that and you go direct to jail
<dneighbo> :)
<Mr_You> software benchmarks?  They would probably make you take down the site, threaten litigation
<reinhard> back
<dneighbo> reinhard we decided geas will do everything
<dneighbo> it will be an operating system, a desktop and cook you dinner
<jcater> oh
<dneighbo> the wipe your butt module will be available later
<jcater> it's the new age emacs then?
<dneighbo> and you will write it :)
mdean (mdean@mkc-65-28-73-63.kc.rr.com) left irc: Ping timeout: 181 seconds
<dneighbo> jcater: key difference emacs doesnt have any 'hygiene' functions
<dneighbo> the author doesnt believe in such things :)
<reinhard> lol
<dneighbo> mdean : DTS crash your system :)
<Mr_You> well gotta do stuff, have a happy thanksgiving!
<dneighbo> thanks
<dneighbo> you too
<dneighbo> guess everyone bailed on the discussion :(
<reinhard> Mr_You: thanks and you too
<reinhard> even i don't have thanksgiving :)
<reinhard> even though
<dneighbo> they dont celebrate the day that the nazi's left?
<dneighbo> or anything?
<reinhard> lol
<reinhard> you sure you wanna know what day we celebrate in austria?
Action: dneighbo thinks austria needs and arnold schwarzenger(sp) day
<dneighbo> reinhard yeah
<reinhard> the day the americans left
<reinhard> :)
<reinhard> oct 26, 1955
<dneighbo> rofl
<dneighbo> thats great
<dneighbo> what is it called?
<dneighbo> oust der bastards
<reinhard> well actually all occupation troups left
<reinhard> us, france, uk, and russia
<reinhard> on that day
<reinhard> it is called "national day"
<reinhard> and austria is still neutral since that day
<reinhard> well btw austria is the only place in the whole world
mdean (~mdean@mkc-65-28-73-63.kc.rr.com) joined #gnuenterprise.
<dneighbo> der Tag lo"schten wir die dummen Scheists
<Mr_You> I have an idea.. it looks like GNU Bayonne is looking to integrate with an XMLRPC app... maybe they are also interested in "RPC" abstraction 
<reinhard> where the ussr put their feet on the ground
<dneighbo> what about switzerland?
<dneighbo> ah
<reinhard> and left without being forced by war
<Mr_You> Some may recall I created Apennine to provide a generic XMLRPC server that can sit on top of all of GNUCOMM, including GNU Bayonne. By using dso modules, one can add functionality for those GNUCOMM services running on each machine. 
<dneighbo> Mr_You yes we have scheduled talk with them for monday 
<Mr_You> word 
<Mr_You> ok gotta run.. will be demoing the 0.1.0 stuff to some family hehe ;-)
<dneighbo> dotgnu, gnue, gnucomm, dcl, phpgroupware are tryin gto have a 'conference'
<dneighbo> of the minds
<Mr_You> conference call?
<dneighbo> yes
<Mr_You> can some lurkers also join?
<dneighbo> though we are looking probably to move it to structured private irc instead
<dneighbo> for cost and other reasons
<dneighbo> i think the idea was
<Mr_You> ahh works for us heh
<dneighbo> have the core heads meet for an hour
<dneighbo> get some stuff knocked out
<Mr_You> why not open moderated IRC?
<dneighbo> then allow anyone from the projects to join in after the fact
<dneighbo> with a back log provided
<dneighbo> um cause none of us have done moderated irc before :)
<Mr_You> ok either way I guess
<nickr> Its easy
<Mr_You> its not hard.. just set /mode #gnue +m ; /mode #gnue <nick> +v
<dneighbo> i suppose you just set a moderator
<dneighbo> kill the other folks
<nickr> you make a channel +m and +v anyone who can talk
<Mr_You> those with +v can talk
<nickr> yea
<dneighbo> and msg the mod when you wish to speak
<dneighbo> hmmm
<dneighbo> thats an idea
<Mr_You> well if you want several with +v and a "moderator" for -v users to msg
<nickr> any o can set +v
<Mr_You> sometims you need to annouce in the channel "msg so and so if you wish to speak"
<dneighbo> right
<Mr_You> anyone?  only ops can change modes
<dneighbo> im assuming at the beginning we dont want open talks (i konw that sounds bad)
<nickr> any +o that is
<dneighbo> but we want just the leadership of the projects to try to set course
<dneighbo> though there is no intent to hide conversation so having observers could be good
<dneighbo> this way the observers could msg the leader for thier projects
<Mr_You> well, if its going to have a public log then it doesn't really matter, I'd just rather see in real time.. you could say the moderator isn't taking msgs 
<dneighbo> if they wish to interject and the lead from that project can choose to include it in discussion or not
<dneighbo> its thought to chew on at least
<dneighbo>  :)
<dneighbo> i think the tenative schedule if for 9pm EST monday
<Mr_You> works for me ;-)
Action: Mr_You not morning person unless yer talking closer to 12am.
<dneighbo> some key people havent confirmed so its DEFINITELY subject to change
ToyMan (~Stuart@c5300-1-ip195.albany.thebiz.net) joined #gnuenterprise.
<Mr_You> have a good one...
Action: Mr_You &
<mdean> dneighbo: somebody picked up maintenance of WISE
jbailey (~jbailey@Toronto-HSE-ppp3639170.sympatico.ca) joined #gnuenterprise.
mdeanlt (~Mdean@mkc-65-28-73-63.kc.rr.com) joined #gnuenterprise.
Remosi (~isomer@210-86-57-16.jetstart.xtra.co.nz) joined #gnuenterprise.
<Remosi> #$@#$
<reinhard> good night all
reinhard (rm@62.47.44.7) left irc: "Omnis enim res, quae dando non deficit, dum habetur et non datur, nondum habetur, quomodo habenda est.  -- Aurelius Augustinus"
<dneighbo> mdean if wise i think you are talking about
<dneighbo> its probably linas vesptas
<dneighbo> of gnu cash
<mdeanlt> nope
<mdeanlt> he handed it off
<nickr> wise?
<dneighbo> which wise are you talking about?
Action: dneighbo might be smoking crack
<mdeanlt> Linas'
<mdeanlt> but he has been looking for a new maintainer for years
<mdeanlt> there's a freshmeat announcement today
<dneighbo> linas didnt maintain wise
<dneighbo> as far as i recall
<dneighbo> he USED it and is in LOVE with it
<mdeanlt> according to his pm page he did
<dneighbo> but didnt maintain it
<dneighbo> he was actively working to get it out to public
<dneighbo> and recently emailed about finally getting hold of the guy
<dneighbo> to put out as free sfotware
<dneighbo> iirc
<dneighbo> at one point i talked to linas almost daily :)
<mdeanlt> the problem as he put it was that it used a broken yacc
<mdeanlt> from his pm page: WISE is no longer actively maintained. I acted as maintainer for several years, and am looking for someone willing to take over this work. 
Action: jbailey is away: I'm busy
<mdeanlt> WISE has a very powerful, very elegant and interesting way of defining workflow, roles, and privileges, and is thus a potentially extremely powerful system. This ability puts it way beyond the capabilities of virtually any other bug tracking system or workflow manager out there. Unfortunately, this feature is implemented through an enhanced yacc (newyacc) that's buggy. WISE should be carefully studied, the workflow concepts carefully, sur
<mdeanlt> gically removed, and then re-written. TBD: remove need for newyacc, update build system, broaden SQL support, modernize cgi infrastructure. 
<dneighbo> hmmmm
<dneighbo> maybe linas is on crack :)
<mdeanlt> heh
<dneighbo> somehow when i looked at this i recall it not being nearly as cool as he thought it was
<dneighbo> but i might have been missign something
<mdeanlt> maybe the backend is good - the frontend sux
<dneighbo> perhaps this is why i was turned off
<dneighbo> i recall thinking wtfk seemed more promising
<mdeanlt> yes
<mdeanlt> I briefly looked at wftk's xml stuff - looks pretty good
<dneighbo> funny, a guy wrote me today, to take over the maintainership of an
<dneighbo> obsolescent CRM package I'd worked on 6 years ago (wise.  wise was cool.
<dneighbo> It makes bugzilla look like a toy). The guy seems eager and capable, 
<dneighbo> we're talking design issues 
<dneighbo> is what linas said on nov 19th this year so like 3 or 4 days ago :)
<dneighbo> guess they found another maintainer in that time frame?
<dneighbo> errr 
<dneighbo> i guess rather that guy is matt (my bad)
<dneighbo> i misread this
<dneighbo> i thought he was saying someone contacted him to take over wise
<dneighbo> as in him take over the wise project
<dneighbo> when in reality someone contacted him to say they would be willing to :)
<dneighbo> i suppose if i would have 'read' my mail :)
<dneighbo> i could have responded to your initial comment hear as
<dneighbo> 'yeah linas told me that 3 days ago' :)
<mdeanlt> :)
<dneighbo> actually i think that mail was to gnucash devel list (me cant keep his mails straight)
<mdeanlt> if anyone cares (since this was bought up recently): Cygwin does include postgresql 7.1.2 as part of the installation options
Action: jbailey is back (gone 00:08:21)
<dneighbo> woo hoo
neilt (~neilt@dhcp64-134-54-175.chan.dca.wayport.net) joined #gnuenterprise.
<jbailey> I should start beeting on that cygwin cross-environment again.  I wonder how far I got to?
<mdean> I'm updating the laptop with it :)
dyfet (dyfet@65.188.113.57) joined #gnuenterprise.
<mdeanlt> dneighbo: http://actionfactory.sourceforge.net/ - doesn't appear very active tho
dyfet (dyfet@65.188.113.57) left irc: "Client Exiting"
<dneighbo> doh that site is so bad i cant use it :)
<mdeanlt> left nav is broken, but the top works
<dneighbo> yeah i noticed that but still painful :)
<dneighbo> time to leave
dneighbo (dneighbo@mail.libertydistribution.com) left irc: "BitchX: to the rescue!"
<dtm> high
chillywilly (~danielb@d104.as7.nwbl0.wi.voyager.net) joined #gnuenterprise.
<jbailey> Heya Penguin-man!
<chillywilly> howdy
Action: chillywilly is already dreaming of a nice turkey dinner with stuffing....mmmmmmmmmm
<chillywilly> why are there 2 mdeans?
<jbailey> That's right, it's American slaughter-the-indians-and-appropriate-culture-day tommorow, isn't it?
<chillywilly> they're Native Americans
<chillywilly> and they weren't slaughtered during thanks giving
<jcater> yes, it's we-will-later-on-slaughter-the-native-americans-and-appropriate-culture-date
<chillywilly> I don't think any Indians were slaughtered by pilgrims...it took civilized aerly america to do that
<chillywilly> ;)
<jbailey> chillywilly: Depends if you consider smallpox as slaughtering...
<chillywilly> jbailey: did you just not celebrate thanksgiving in Canada a while ago?
<jbailey> chillywilly: Yeah.  I'm not really clear on the whole point of Canadian thanksgiving.  I had originally learned that it was just 'cause people were thankful to get off the boat, but there's no shortage of cultural appopriation...
Action: chillywilly misses his family and is glad for a time to be able to get together ;), I could care less what happened 1000 years or so ago
<chillywilly> I can't do anything about that
<chillywilly> plus the Packers are playing the Lions in the super dome ;)
<chillywilly> can't have thanksgiving without some football
<chillywilly> ugh, I hope they don't have Madden and Sumerall doing the thanksgiving game
Action: chillywilly is away: dinner
Action: jbailey is away: shopping
Action: chillywilly is back (gone 00:13:47)
neilt (neilt@dhcp64-134-54-175.chan.dca.wayport.net) left irc: "later all"
ra3vat (ds@195.239.64.51) left irc: "[x]chat"
Action: jbailey is back (gone 01:26:42)
jbailey (jbailey@Toronto-HSE-ppp3639170.sympatico.ca) left irc: "Client Exiting"
Maniac (~darryl@h24-82-132-209.wp.shawcable.net) joined #gnuenterprise.
<chillywilly> who changed the look of the web site?
<chillywilly> looks better :)
<jcater> cool
<jcater> thanks
<jcater> you're the first person to give me feedback
<chillywilly> hmmmm
<chillywilly> how do I change my theme to match that one?
<chillywilly> er, which one is that?
<chillywilly> when I login I get the old one
<chillywilly> GNUeSlashOcean?
<jcater> yes
<chillywilly> yep, looks better now ;)
<chillywilly> anyone know how to do some gimpage?
<chillywilly> I add dyntext, but how can I move that...I am tryint to make a logo
<chillywilly> it is a new layer (the dyntext)
ToyMan (Stuart@c5300-1-ip195.albany.thebiz.net) left irc: "Client Exiting"
<derek> jcater: not ture i gave you feedback earlier
<derek> i like parts better
<derek> and dislike other parts 
<derek> for those asleep at the wheel we are on linuxtoday and enterprise linux today :)
<derek> for our v0.1.0 release
<chillywilly> yep
<chillywilly> I saw that
<derek> holy crap
<derek> you realize that the freshmeat release, plus kernel cousin, plus eltoday story has at over 75,000 hits today :)
<chillywilly> woah
<chillywilly> time for some advertising money
<chillywilly> :P
<Remosi> heh
<Remosi> who needs advertising?
<chillywilly> we can advertise on the site :P
<chillywilly> I am joking anyway
<chillywilly> get with the program
<Remosi> hmm
<Remosi> I'm in for a cut of the clickthru's :)
<chillywilly> must have contributed GNUe CODE ;)
<chillywilly> or code
<Remosi> Dang :P
jcater_ (~jason@HubA-mcr-24-165-193-222.midsouth.rr.com) joined #gnuenterprise.
Action: Remosi goes on a bug hunt so I can submit a patch soon
<jcater_> rock!
<jcater_> front page of linux today
<jcater_> wtf
<jcater_> jcater go away
jcater (jason@w202.z065105010.mem-tn.dsl.cnc.net) left irc: 
<jcater_> thank you
Nick change: jcater_ -> jcater
Maniac (darryl@h24-82-132-209.wp.shawcable.net) left irc: Remote closed the connection
Maniac (~darryl@h24-82-132-209.wp.shawcable.net) joined #gnuenterprise.
alexey (alex@techkran.vladimir.ru) joined #gnuenterprise.
jcater (jason@HubA-mcr-24-165-193-222.midsouth.rr.com) left irc: "beddy time"
chillywilly (danielb@d104.as7.nwbl0.wi.voyager.net) left irc: 
--- Thu Nov 22 2001