*** johannesV has joined #gnuenterprise *** btami has joined #gnuenterprise *** reinhard has joined #gnuenterprise *** kilo has joined #gnuenterprise *** chillywi1ly has joined #gnuenterprise mornin' *** chillywilly has quit IRC bah good morning frickin' IP changed and now my ssh session is slow as molasses morning chillywilly: goot articles from W.Ambler yesterday, thx afain s/afain/again good * btami can't type early morning you can type early morning just like this: "early morning" 8-)) btami: yea that site rocks wow, major lag now to the server at work... *sigh* running a traceroute and it is not looking pretty derek have to read that :) I love it when it does: 13 * * * 14 * * * BAH oh I bet it is because they block ICMP packets yes our ISP at the office does that hmmm still ssh is really painful and slow :( was workign fine earlier man this blows *** chillywi1ly has joined #gnuenterprise *** chillywilly has quit IRC *** dcmwai has quit IRC *** dcmwai has joined #gnuenterprise *** SachaS has joined #gnuenterprise good morning morning SachaS hi kilo hi SachaS: please look at http://www.agiledata.org/ (if you havn't lookd at) and check emails *** dimas_ has quit IRC *** dcmwai has quit IRC *** SachaS has quit IRC *** SachaS has joined #gnuenterprise *** dimas has joined #gnuenterprise *** reinhard has quit IRC *** reinhard has joined #gnuenterprise *** btami has quit IRC *** kilo has quit IRC *** reinhard has quit IRC *** ajmitch has quit IRC *** dsmith has joined #gnuenterprise Good Morning, #gnuenterprise Is svn broke again? *** Morphous has joined #gnuenterprise *** jamest has joined #gnuenterprise hi dsmith *** Amorphous has quit IRC SachaS: Greetings *** Amorphous has joined #gnuenterprise *** johannesV_ has joined #gnuenterprise *** reinhard has joined #gnuenterprise *** wayneg has joined #gnuenterprise *** Morphous has quit IRC *** johannesV has quit IRC servus servus austria ;) grüezi ;-) *** jcater has quit IRC reinhard: you saw the abstraction of my diagram ... another abstraction level will be added :) one on top of the bus. its no longer simply publish/subscribe but its a business process on top of publish/subscribe. there is a collaborative business process handled in the ebXML software there is a business business process handled in the business application (gnue appserver app with a possibly public procedure which uses business classes and their procedures) but then there is a new business process a business process describing how applications interact with each other, this might be called a integration business process. derek have to read that :) im not sure what that is about for the record i am probably one of the few here that has made a commerical application that actually used a purely object database so telling me to look at the advantages of objects is silly you act as if i dont know what objects are or don't understand they can be valuable i will argue teh OTHER way what REAL experience do you have on an enterprise level with object data storage? i can state i built and maintained commercial ERP software for major law firms using object database technology so i know some of the real advantages and real pitfalls not "book" knowledge, but real practical implementation for the record http://www.omegalegal.com/ is the software i am talking about *** jcater has joined #gnuenterprise *** dhill1 has joined #gnuenterprise jcater or jamest is there a way to force focus on an entry? jamest in GFEntry what did I win? i dont see in triggers or properties to set such a thing basically looking for something like I think it's just setFocus(entryName) blk.entry.setFocus or form.setFocus(entryName) hmm okay ah so its probably in GFForm yip GFForm GFForm.py: self._triggerFunctions = {'setFocus':{'function':self.tri thanks was looking in wrong area do we have a "tool" to whip through all the self._triggerFunctions *** btami has joined #gnuenterprise and print out documentation on that? as seems like 40% of the questions the guys have here (and me too) is how do i do X no been thinking about that but haven't had a chance to do it we are getting better at looking the source for trigger definitions have you looked in the Dev Guide? no well maybe i had the guys pritn out something but they complained it was missing a lot might have been an old version you have link to most current? it is missing a lot i will download and print but it's also 76 pages http://www.gnuenterprise.org/tools/forms/docs/Developers-Guide.pdf one thing we want to do is hep document the stuff that is there is pretty complete and up-to-date Appendix C is what you want * jcater may try to spend a few minutes today updating that printing it now derek have to read that :) http://www.agiledata.org/essays/mappingObjects.html http://www.ambysoft.com/persistenceLayer.pdf jcater: hav you got a little time today? well, I don't have time to give my opinion but I have a strong opinion about all of this probably not :( * jcater is about 2 weeks behind schedule on my current project btami: this guy explains exactly my PROBLEMS with the approach There is an impedance mismatch between the object and relational paradigms, and this mismatch leads object and data practitioners to different designs. basically he argues build all the classes and such and make the database structure to mimic those i.e. lose all the power of the relational database storage system you are using he tries to pass this off as "anyone that is not pro object is an idiot and they need to change how they think" if they understood objects they would design their databases to mimic them when in fact peopel that understand what relational design was designed to solve take advantage of it and really do not want to restrucutre their data to be stored like objects as they lose the features they inherently built in to take advantage of teh system clearly if one wants to say "build your classes and make your storage match" why bother using relational database system? save the headache and use an object or hierarchial storage system that has an "odbc" driver that allows you to map your classes to sql tables for other applications to "read" the data i do think mapping can (and should exist) that in my opinion is WHAT appserver should be doing i do not think we should be making whacked schemas to match classes that is my opinion (and like standards that great thing about opinions are there are so many to choose from) derek: maybe you are reading another article :) i can't read nowhere he tries to pass this off as "anyone that is not pro object is an idiot and they need to change how they think" as i understand, these papers all about how to use OO languages/design _with_ rdbms and Python is an OO language... our DBSIG2 abstraction is a mapping too appserver is another one just enhanced with procedures maybe i'm wrong *** bluesbaron_ has left #gnuenterprise *** Amorphous has quit IRC * btami thinks derek have to read the gnue-appserver/doc/whitepaper.txt again btami my point is he says CLEARLY dont use traditional relational structures they are bad make them match you classes http://www.agiledata.org/essays/drivingForces.html he starts by attackign that relational databases are all most always poorly designed "First, your legacy database design likely has some significant problems." then he adds even if it is not designed badly it still wont work "Second, even if the existing database design is very good there can be significant differences in the way that you map objects to relational databases." im all for mapping and i think site actually has some good points and is valuable however on the whole they advocate losing relational functionality in favor of making the mapping easier as evidence by the line i just pasted "even if the relational design is good" "it might not be good because it doesnt map well to objects" that is how i read it and the previous two times we have written appserver this has proven to be the case objects get discussed then we start going down the class path (no pun intended) at which point in time we decide to map to relational tables and soon it turns into extending the relational tables to account for object needs then it turns into altering relational tables completely to make mappings easier derek: stop and then it turns into the "drivers" for forms don't work well anymore have you read whitepaper.txt ? and then the people doing the work on appserver get pissed off an leave because they really want an object database and it just took them a while to realize it is it good or bad for your needs? btami: im expressing the pattern not saying that is the goal it is good.. i like the ideas in the whitepaper (if it is the same one i read many moons ago) my point being you are pointing me to a site say ing "read this and then you will be enlightened" and im saying i'v never sait this !!! "what you want me to be enlightened on is exactly what has killed appserver development in the past" i know nothing the old appserver derek have to read that :) i'm trying to use the new :) so you didn't want me to read it to be enlightened? im not asking you to use the old no, just read :) im saying "don't forget history" it has a way of repeating itself if you agree on the whitepaper, what is you against with the current implementation? i think it fallows the whitepaper im not against the current implementation someone is telling you lies if they have said such a thing this conversation started out of the "gcd" discussion i simply asked "why" a gcd and how does the gcd replace and/or compliment the gsd is a gcd necessary hmm i think yes and instantly i was attacked as "not understanding objects" and "hating the appserver" the discussion on gcd's had me worried we were ditching sound relational structures in favor of makign structures that were "easy to map" i merely raised the flag to be cautious and now somehow "i hate appserver" ok apparently you can't question anything here or you hate the product? i think the .gcd problem is as much as i give jamest and jcater crap about forms.. (read old logs with me telling them i hate blocks) what format will be used for formalise gnue-packages? *** uruboro has joined #gnuenterprise you would be astonished that im trying like hell to get forms used a major setting for a major application point: there has to be open discussion with out getting personally offended because i question the validity of something doesnt mean its the wrong thing (i.e. i have come to actually like blocks some now that they are being used (when i couldnt see their use initially)) sure and because someone asks questions doesnt mean they dont like the product you will have to excuse my frustration ok * derek is pretty adamantly against "objects" i want an app server not an object server there has been much debate (and likely at some point either this project or appserver will fork over this issue) this is what you said btw but i kind of feel like you, kilo and reinhard have turned this into a personal attack that i hate appserver and as you just pasted i WANT appserver and yes have "concerns" about objects just to remind you but not remember we are using the term object in different ways btami has asked you what you think of gcds and you answered that likely "appserver or this project will fork" note that appserver is not part of this projects project it's appserver on the one side and "this project" on the other side the comment about the fork is that once already a fork has occured over this issue *** wendall911 has joined #gnuenterprise * reinhard will shut up again now reinhard: my point on the fork is that either gnue could end up with 2 appservers if there is fundamental disagreement or if someone didnt like forms being relational.. they could fork the project as a whole * derek did not say this would happen or that it was desired to happen in fact it would be nice to be avoided at all costs just pointed out that it could likely happen if the debate is intense as it has happened once example.... treshna and bond-db http://freshmeat.net/projects/bonddb/ i like the whitepaper and appserver very much i am itching to be able to use it here im not opposed to gcd's, but i think deciding on them will bring on great debate i htink at some point we need to schedule an irc meeting to hash out how to handle class definitions and module packing as btami has a very valid issue that its hard to make appserver packages without some strucutre like that ok, just for log i like all you guys :) really rofl they probably don't have budweiser commericals in hungary, but... "I love you man!" followed by "You aren't getting my bud light" would be an appropriate joke here half the audience gets the joke, the other half scratch their heads. derek, have you tried (within last 3 months) the sample in appserver? i like SachaS too :) thanks btami i like you too haha :))) i like SachaS especially he has done more for appserver lately than probably any other except those getting paid for it thanks reinhard and btami for getting this stuff to run under windows and jamest and jcater for their great work on forms, reports, and common and derek actually i liked siesel too for taking the blame for everything :) just for log again, i hate win :) he had an open ear on mostly everything i said i like johannesV_ commits :) oh sorry yes of course i like johannesV_, too :) now could someone please pass the joint to me? :-P rofl ok,please stop, enough love here :) errm, just svn needs some more love, hey jamest! oh i like reinhard very much by the way * SachaS going to the movies, needs a break. "Day after tomorrow" *** uruboro has quit IRC hmm did someone pass some weed around that I missed? seems like a hippie love fest in here ummmm I will kick all your asses is that better jcater ? oooo kinky, eh troutslap == gnue slang for take a hit off the doobie and pass it on *** johannesV_ has quit IRC *** Amorphous has joined #gnuenterprise *** sjc has joined #gnuenterprise *** dhill1 has quit IRC jamest: Heh. Did svn hoze again? Did you disable that killall? s/Heh/Hey/ the killall wasn't a problem OK. wassup dale? man I am tired * chillywilly sleeps on his keyboard * dsmith hands cw a mountain dew * dsmith waits for qt to compile again evil ;P might as well run Gentoo ;P oh wait that was bddebian who was calling me a heretic s/compile/cross-compile/ nevermin d oh ye a mr. frickin' embedded systems you suck I want to do embedded linux embed this! We are looking for two people. where are you located? Cleveland Ohio hmmm what kinda stuff are you working on? 410k hmm? 401k? chillywilly: Automotive scan tools. OBDII stuff OBD II stuff? ok cool a friend works for autologic company well havoc is doing some work there too but they have to use VB Ewwww because of the PDA manufacturer We were going to use mips (Alchemy) because of some costly manufacturng issues, we are now looking at powerpc. cool s/because/but because/ so how low level do you go with it? you get to use C or what? So I have another new toy to play with. C Device Drivers. how many device drivers did you write before you started working there/ On Linux? None. HEH But I've fexed several Alsa mostly oh dude I've written dos and os/2 drivers, and I've done plenty of embedded systems with no os. Phil Keaggy was in town yesterday Cool. My Man I was lame and didn't go see him though My Phil. s/My/Mr/ I have a home project I'm toying with. what is that? A tape looper kind of box I see I got a webpal for cheap. k With another ISA soundcard, I think I can turn it into a jamman like device. Not enough time though man I am tired And then sooperlooper just came into Debian, and it's inspiring me again. chillywilly: Get to Work! I am I have a web support issues now *** btami has quit IRC chillywilly: You know I went to the very first Glass Harp reunion concert? ok Ya it was great. He wrote a song about it. "Wish you were there". *** kilo has joined #gnuenterprise *** reinhard has quit IRC hi kilo hi SachaS how are you? angry a little bit why is that? yeek installing windows on one machine, reparing some fscking HDD on the other, and my 2nd favorite keyboard is completely silly... and i have just read the irc log... oh ok. are you already modelling your hr application? on paper ... maybe well, work has begun already use case diagrams and class diagrams are built and being refined over and over ok. hmmm. ok. you found useful information in the old .gcd files? yes, there are some. i just can't see why this debate is on... you played with classrepo.gfd. I also did. so we two are VERY much aware that something has to be done... i think the ones who are actually working with appserver see the issue from a user stand point. and let's agree from now on that we don't use the word 'object', but say... emm.... errr... 'blopp' instead 8-))) tcejbo how do you prononuce that? dunno Can't pronounce it, keeps people from yelling. or it could be 'blap', so you can slap with a blap tseebo Everyone knows that objects are backwards, anyway. haha * dsmith runs and hides what do you mean with "objects are backwards" dsmith? that tcejbo is object written backward or do you mean something else? SachaS: Ya, just a stupid joke/idea SachaS: TO sort of go along with the objective discussions going on lately ok. *** sjc has quit IRC *** dsmith has quit IRC SachaS: could you please mail me today's mailings on subject gcd? can't get hold of them lost in space they are ok. sent thx perfect. just tumbled over my son's ball... just like an elephant i was... perfect day *** reinhard has joined #gnuenterprise *** jamest has quit IRC servus reinhard. cant you sleep? servus reinhard, servus austria hi off to sleep now *** reinhard has quit IRC story: Open source on offense in ERP and business applications market http://software.newsforge.com/software/04/05/22/193215.shtml quit sleeping on keyboard *** kilo has quit IRC *** dhill1 has joined #gnuenterprise *** jcater has quit IRC