I guess that's what they mean by a task management tool - 'a friend of mine said "I don't remember what's happened in the last two days. But according to DCL, it was very profitable."'
Table Of Contents
|1.||21�Feb�2003||GNUe Small Business overview|
|2.||23�Feb�2003||Breaking CVS HEAD to add new features|
|3.||24�Feb�2003||Persistance and OOP with Application Server|
|4.||24�Feb�2003||Reporting bugs with GNUe tools|
IntroductionThis Cousin covers the three main mailing lists for the GNU Enterprise (http://www.gnuenterprise.org) project, plus the #gnuenterprise IRC channel. GNUe Traffic is now group-authored: if you'd like to join the team, let us know (mailto:firstname.lastname@example.org)
1. GNUe Small Business overview
21�Feb�2003�Archive Link: "[IRC] 21 Feb 2003"
Summary By Arturas Kriukovas
Topics: Small Business
People: Mike Vincent,�Derek Neighbors,�Stuart Quimby,�Jason Cater,�James Thompson
Derek Neighbors (revDeke) said he was hoping to get moving on gnue-sb. Mike Vincent (Vee2d2) asked maybe he had "some notes or something for what you have in mind for gnue-sb?" Derek admited he doesn't - "the general gist of it is this - what a small business needs to operate done in gnue two tier. I would imagine a quickbooks replacement with more funcationality original needs were for someone else at had start in inventory/warehouse management. The other need was contact management thats why you see contact manager and item stuff in gnue-sb today. We were going to try to 'port' NOLA to gnue-sb for accounting as a start, though im not as convinced on that now, but its still possible. I dont want to build for buildings sake, i want requirements from someone who plans on using it. Having hard time finding someone who doesnt want something like yesterday, i.e. willing to go through a development process or who is willing to go through a development process and actually willling to commit time, i.e. we have had like 10 opportunities at this with different people and we usually get one of two things happening: 1. We dont do this full time so a solution wont be ready in 2 weeks. So they go elsewhere. 2. They don't have any time to put towards letting us know their needs so we can make something for thme."
Later, Stuart Quimby (ToyMan) said that he was now working with Derek on specifying the requirements of his educational toys (http://www.dstoys.com/) business for GNUe Small Business - "almost ready to be the bleeding edge poster-child" . Derek noted that gnue-sb was intended to be a generic solution for most small businesses, but by being involved from the start, potential users could help ensure it "should be suitable to your needs (thats a plus of going first) ;)" He was "seriously ready to start bolting some stuff up tonight" - "i would like to work with you (and stu) to roadmap gnue-sb - especially short term plans and get cooking" , starting with a structure for Stock-Keeping Units (SKUs).
Later, Mike asked for the gnue-sb story. So Derek told it - "We were going to do work for a small manufacturer that was in a bad pinch. We couldnt wait for appserver and 'official submission' processes so we made a gnue-sb, that would be less 'controlled' but quicker to build with 2 tier only no appserver. The original needs were item management and inventory as the upfront need was warehouse management basically everything after the order was taken, i.e. they were using paypal and online store to get orders. They wanted to dump an order into system and basically have pick lists created from it for the warehouse bundle ship and have all shipping labels etc printed. Once the backend was working, redo shopping cart side as well as add invoicing and other things. This is of course a light speed overview. Anyhow they got impatient and have now gone through about 4 off the shelf products and still searching."
He had continued to develop gnue-sb anyway, and "The item basics are there (though the sku category stuff needs to be added) it wouldnt be too hard to get basic inventory secondarily. IDEALLY gnue-sb will be perfect for people that use paper today as the transition can be slow and they can use immediately even if functionality is more minimal. One problem you run into 'starting' is people not wanting to 'use' something until it has EVERY feature but its hard to get features when no one will use it :) Im REALLY hoping that you might be in a position to start using a piece at a time for example if we got it so just all your products(items) were able to be entered that is a huge start and if nothing else you can query your product line and run reports on what products you have to offer but then we can bring inventory online or ordering or shipping etc." Mike confirmed that this was his plan. Derek thought "if that makes sense, so say you get products and then get customers or vendors. You can use them even if you are still doing invoices in word or whatever. I suspect if you use this you will be amazed at progress as two things will happen" Firstly, development of the GNUe tools such as Forms and Reports would increase as the tools developers such as Jason Cater (jcater) and James Thompson (jamest) would find it easier to "respond to 'production problems' really quickly as they can relate, plus its encouraging for them to see it in production. 2. putting something in production gets features and bugs fixed quickly :) as motivation all around is bumped exponentionally ;)"
Jason asked whether Michael was doing invoices from a word processor using mail merge. "'cause you know we can do merges between word/oo documents and database data" - both James and Jason used this extensively. "my users can edit their basic documents in word/oo, and save to a "public" drive - so I don't have to keep modifying their reports for them :)" . Derek suggested "we get his products in db and get him using the contact manager - and he can at least 'fudge' invoices for a small while until we get an order screen up (real quick) - and have it make openoffice invoices with reports" . He told Michael "i really think within 2 or 3 weeks we can get you doing invoices in gnue if you are willing to start simple and grow with it" . Derek and Michael talked some more about the practicalities of setting up multi-part Stock-Keeping Units, as previously discussed in Issue�#69, Section�#3� (16�Feb�2003:�Stock Keeping Units) .
2. Breaking CVS HEAD to add new features
23�Feb�2003�Archive Link: "[IRC] 23 Feb 2003"
Summary By Arturas Kriukovas
Topics: Forms, Navigator
People: James Thompson,�Christian Selig,�Jason Cater
Christian Selig (lupo) asked what had changed in cvs during the last 4 weeks. James Thompson (jamest) explained that cvs head was broken (at least forms and designer were unsuable in it's current state). "All forms are broken as I've migrating all popup dialogs into .gfd files and I got sick last week and couldn't complete, won't get back to it till next week" said James. This broke all forms because stuff like "About" box, "Jump to record" box, error boxes no longer worked. Currently James was adding <dialog> support into forms "which is the equivalant of <form style="dialog"> which means <form> now can nest. You'll be able to create custom prompts for your users and call them via triggers." Christian's comment was short - "you're a crazy pervert and you have my deepest respect for nested forms which is indeed a great accomplishment" .
James also noted that Jason Cater (jcater) was "hard at work on" a User Interface (UI) driver that used the qt UI directly, instead of via wxPython. He believed Jason "has the forms displaying (i think)" but "it's not handling events yet I believe" . Christian also asked about Navigator. James said "ooooo, it may have broke too :) - i didn't test it w/ the ui changes :)" Christian asked about Role Based Access Control (RBAC) in Navigator. James said he had not really thought about how to do that yet - "can't speak for others" .
3. Persistance and OOP with Application Server
24�Feb�2003�Archive Link: "[IRC] 24 Feb 2003"
Summary By Arturas Kriukovas
Topics: Application Server
People: Reinhard M�ller
It was asked whether object in appserver could be persistent using a relational database. Reinhard M�ller (reinhard) explained that all objects are persistent when using appserver. "In fact an object in appserver is more or less a database entry and a class is more or less a database table" . It was asked whether it was possible to implement inheritance. Reinhard answered that "we very much believe that we shouldn't do inheritance (we == some of the core project members). What we have as appserver "object" is not in the sense of OOP. We just call it object because it combines data (from the database) with code (python triggers)" . There were some similarities to OOP (private and public properties) - " there are properties with different "permissions", like property "stockCount" is read/write for the "system" but read only for the user so the user may not do stockCount=17, but he may use procedures like "increaseStock(5)" which increases stockCount by 5 but also logs into another table who did it etc."
4. Reporting bugs with GNUe tools
24�Feb�2003�Archive Link: "[IRC] 26 Feb 2003"
Summary By Arturas Kriukovas
People: Derek Neighbors
Derek Neighbors (revDeke) informed that in case a bug is found, standard mode of operation was to come to irc "and see if it can be fixed immediately :)" . In case there were no solutions to the problem, it was possible to email the question to email@example.com (for designer: firstname.lastname@example.org, for forms: email@example.com). "Be sure to give information in your email about version of python, wx, operating system (distro) as well as snippet of the trace. Our support structure is pretty brutalized right now with 0.5.0 development eating a lot of time so it might be highly likely that 0.5.0 will fix this (might be the answer), but its worth submitting none the less" .
He also noted that wx windows 2.4 did not go well with 0.4.3 version of GNUe, so it was better to use 2.2.
Sharon And Joy
Kernel Traffic is grateful to be developed on a computer donated by Professor Greg Benson and Professor Allan Cruse in the Department of Computer Science at the University of San Francisco. This is the same department that invented FlashMob Computing. Kernel Traffic is hosted by the generous folks at kernel.org. All pages on this site are copyright their original authors, and distributed under the terms of the GNU General Public License version 2.0.