You can subscribe to this list here.
2002 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
(26) |
Oct
(39) |
Nov
(1) |
Dec
|
---|
From: Derek H. <lor...@ms...> - 2002-11-18 03:03:04
|
Here's my proposal for structuring the design documentation. As a foundation to base this proposed format on, I've loosely followed the format and structuring of a Design Pattern (from the Gang of Four book). Unlike a design pattern, there is no intention to be general, but rather specific (in particular, progressively less general and more specific the deeper a design goes). Right now, Use Case documents are named U##.xml or A##.xml. My recommendation is that Design documents ("programming specifications") should be prefixed with PRG (for "PRoGramming"). As the extension will remain .xml, and not .cpp, .bas, or .java, these are still documentation and not implementation source files. Following the PRG prefix is a #### number. To allow for several hierarchical levels of abstraction (a high-level design, a low-level design of multiple classes, and a detailed-level design of the intricacies of a single class), we should allocate thousands to functional aspects of eDocs' behavior. For instance, 1000 Search Engine : : 9000 eDocs-Microsoft Integration A high-level Microsoft Integration Design document (perhaps describing the overall web services architecture by which Microsoft Integration takes place, relying on lower-level Design documentation to focus in on narrow areas) would be named PRG9000.XML To take this another level deeper, we may create the following Design documents covering the design of the Web Services interoperability: PRG9000.XML Web Services Integration : Architectural Overview PRG9010.XML Design of the WSDL Interface PRG9020.XML User Authentication : HTTP Cookie? SOAP Header Extension? Session ID argument with each SOAP Call? PRG9030.XML User Authorization : XML Representation of eDocs Roles and Privileges PRG9040.XML Encoding : File Transfer via BASE64 encoding ; encryption? : : PRG9100.XML eDocs-Microsoft Integration : Architectural Overview. PRG9110.XML Microsoft SOAP Toolkit 2 : Specific configuration and technical issues PRG9120.XML Apache SOAP 2.3 : Specific configuration and technical issues PRG9200.XML Client-side COM Component : IDL interface, ATL responsibilities, deployment PRG9210.XML C++ Design of IEDocsCredentials Implementation PRG9220.XML C++ Design of IEDocsNavigator Implementation PRG9230.XML C++ Design of IEDocsLocalRepository Implementation PRG9240.XML C++ Design of ISupportErrorInfo Implementation for extended error information : : PRG9300.XML Microsoft Word Integration : Add-in Model and Considerations for using SOAPClient from VBA PRG9310.XML Design of the Log-In Form : user interface and interactivity PRG9320.XML Design of the Check-In Dialog : user interface and interactivity : : PRG9900.XML Microsoft Windows Explorer Integration PRG9910.XML Design of a Namespace Extension PRG9920.XML Design of Explorer User Interface There's nothing that inherently prohibits cross-references between design documents. For example, it's entirely possible PRG9400.XML may describe Excel Integration (so that Excel spreadsheets can be managed by eDocs), and instead of duplicating the content of PRG9310.XML, PRG9320.XML, etc., it will simply refer to these design documents (provided design content is largely equivalent between Word and Excel contexts, the design is reusable). For the template of the XML, currently I'm working with something like the following (you can ignore the {$*} tags, they're just for a generator I'll use to export design specs): - - - ProgSpecTemplate.xml <?xml version="1.0" encoding="iso-8859-1" ?> <!-- Copyright notice and GPL statement go here --> <document url="./{$ID_LOWERCASE}.xml"> <properties> {$BEGIN_LOOP} <author>{$AUTHOR}</author> {$END_LOOP} <title>{$ID_UPPERCASE}. {$TITLE}</title> <previous name="{$PREVIOUS_SECTION} Index" href="../{$PREVIOUS_SECTION_LOWERCASE}_index.html" /> </properties> <body xml:lang="en-US" > <section name="Brief Description" href="toc"> {$BRIEF_DESCRIPTION_HTML} </section> <section name="Detailed view" href="toc"> <detview-section name="Characteristic Information" > <detview-section-item name="Goal">{$INTENT_PURPOSE_HTML}</detview-section-item> <detview-section-item name="Scope">{$SCOPE_HTML}</detview-section-item> <detview-section-item name="Rationale">{$INTENT_RATIONALE_HTML}</detview-section-item> <detview-section-item name="Consequences"> {$CONSEQUENCES_RESULTS_HTML} {$CONSEQUENCES_TRADEOFFS_HTML} </detview-section-item> <detview-section-item name="Related Use Cases"> <ul> {$BEGIN_LOOP} <li><a href="{$RELATED_USE_CASE_ID}.html">{$RELATED_USE_CASE_ID}.</a> {$RELATED_USE_CASE_TITLE}.</li> {$END_LOOP} </ul> </detview-section-item> <detview-section-item name="Future Extensions"> <table> {$BEGIN_LOOP} <tr> <td>{$LIKELY_FUTURE_EXTENSION_HTML}</td> <td width="10%">{$LIKELY_FUTURE_EXTENSION_SCOPE}</td> <td width="10%">{$LIKELY_FUTURE_EXTENSION_LIKELIHOOD}%</td> </tr> {$END_LOOP} </table> </detview-section-item> <detview-section-item name="SAAM Score">{$SAAM_SCORE}</detview-section-item> </detview-section> <detview-section name="Structure"> <detview-section-item name="Architecture">{$ARCHITECTURE_HTML}</detview-section-item> <detview-section-item name="Data Dictionary"> <table> <th> <td>TERM</td> <td>TYPE</td> <td>WIDTH</td> <td>NULL?</td> </th> {$BEGIN_LOOP} <tr> <td class="ddterm">{$DATA_DICTIONARY_TERM}</td> <td class="ddtype">{$DATA_DICTIONARY_DATATYPE}</td> <td class="ddlen">{$DATA_DICTIONARY_WIDTH}</td> <td class="ddnil">{$DATA_DICTIONARY_NULLABLE}</td> </tr> <tr> <td class="dddef" colspan="4">{$DATA_DICTIONARY_DEFINITION}</td> </tr> {$END_LOOP} </table> </detview-section-item> <detview-section-item name="CRC"> <table> <th> <td>CLASS</td> <td>RESPONSIBILITIES</td> <td>COLLABORATORS</td> </th> <tr> {$BEGIN_LOOP} <td><b>{$CLASS_NAME}</b></td> <td>{$CLASS_RESPONSIBILITIES_HTML}</td> <td>{$CLASS_COLLABORATORS_HTML}</td> {$END_LOOP} </tr> </table> </detview-section-item> </detview-section> <detview-section name="Implementation"> <detview-section-item name="Techniques">{$IMPLEMENTATION_TECHNIQUES_HTML}</detview-section-item> <detview-section-item name="Pitfalls">{$IMPLEMENTATION_PITFALLS_HTML}</detview-section-item> <detview-section-item name="Technology-specific Issues"> <ol type="number"> {$BEGIN_LOOP} <li> <h3 class="imphdr_q">{$IMPLEMENTATION_TECHNOLOGY_ISSUE_QUESTION}</h3> {$IMPLEMENTATION_TECHNOLOGY_ISSUE_ANSWER_HTML} </li> {$END_LOOP} </ol> </detview-section-item> <detview-section-item name="Platform-specific Issues"> <ol type="number"> {$BEGIN_LOOP} <li> <h3 class="imphdr_q">{$IMPLEMENTATION_PLATFORM_ISSUE_QUESTION}</h3> {$IMPLEMENTATION_PLATFORM_ISSUE_ANSWER_HTML} </li> {$END_LOOP} </ol> </detview-section-item> <detview-section-item name="Language-specific Issues"> <ol type="number"> {$BEGIN_LOOP} <li> <h3 class="imphdr_q">{$IMPLEMENTATION_LANGUAGE_ISSUE_QUESTION}</h3> {$IMPLEMENTATION_LANGUAGE_ISSUE_ANSWER_HTML} </li> {$END_LOOP} </ol> </detview-section-item> </detview-section> </section> <section name="Related Information" href="toc"> <detview-section name="Sample Code (fragments)"> {$BEGIN_LOOP} <a name="SAMPLE_{$ID_UPPERCASE}_{$IDX}"> <sampleCode relPath="{$SAMPLE_CODE_RELPATH_SPEC}" static="{$SAMPLE_CODE_STATIC}" start="{$SAMPLE_CODE_START_LINE_NUM}" end="{$SAMPLE_CODE_END_LINE_NUM}" alt="{$SAMPLE_CODE_CAPTION}" /> </a> {$END_LOOP} </detview-section> <detview-section name="Class Diagrams"> {$BEGIN_LOOP} <a name="CLASS_{$ID_UPPERCASE}_{$IDX}"><img src="./diags/class/{$ID_UPPERCASE}_c{$IDX}.png" alt="Class Diagram {$IDX}." /></a> {$END_LOOP} </detview-section> <detview-section name="Sequence Diagrams"> {$BEGIN_LOOP} <a name="SEQ_{$ID_UPPERCASE}_{$IDX}"><img src="./diags/seq/{$ID_UPPERCASE}_i{$IDX}.png" alt="Sequence Diagram {$IDX}." /></a> {$END_LOOP} </detview-section> <detview-section name="Statechart Diagrams"> {$BEGIN_LOOP} <a name="STD_{$ID_UPPERCASE}_{$IDX}"><img src="./diags/std/{$ID_UPPERCASE}_s{$IDX}.png" alt="State-Transition Diagram {$IDX}." /></a> {$END_LOOP} </detview-section> <detview-section name="Deployment Diagrams"> {$BEGIN_LOOP} <a name="DEPLOY_{$ID_UPPERCASE}_{$IDX}"><img src="./diags/deploy/{$ID_UPPERCASE}_d{$IDX}.png" alt="Deployment Diagram {$IDX}." /></a> {$END_LOOP} </detview-section> <detview-section name="Screen Layouts"> {$BEGIN_LOOP} <a name="SCREEN_{$ID_UPPERCASE}_{$IDX}"><img src="./diags/screen/{$ID_UPPERCASE}_fig_{$IDX}.png" alt="Screenshot: {$IDX}" /></a> {$END_LOOP} </detview-section> </section> </body> </document> - - - Does anyone have recommendations or counterproposals for which 'sections' should be present in the design documentation? Derek Harmon |
From: Sergio R. <sra...@ti...> - 2002-10-31 12:56:10
|
Piotr, ok for you, unfortunately, because of we're in the very early stages of development nothing has been assigned to you right now. You'll be really involved later. In the meanwhile, if looke at the design steps you can begin to see some sample screens proposals. Look at the security section of the analysis documentation. You've done a good work with the web site. I think we can wait to register it on a searchengine until we'll be in a more mature phase of the product. Regards Sergio ----- Original Message ----- From: "pk" <pkr...@wp...> To: "edocs development" <edo...@li...> Sent: Thursday, October 31, 2002 12:34 PM Subject: Re: [Edocs-development] Status update > On 10/29/2002 11:43 PM, Uz.ytkownik Sergio Ramazzina napisa?: > > Hi to everybody, > > > > I appreciate to have a status update on the activities that everybody has > > been assigned to. > > Can I hope to have something to discuss for the first week of November? > > On my part: > > - website is up and running; It has the features You asked for and is up > to date. BTW, I have one question - should I register our website with > the searchengines? I wasn't sure about that because we're still in early > stages of the development so I'm not sure if it should be done now. > > - user interface - we have agreed on the layout some time ago so I > thought we're through with it. About using struts new Tiles feature - I > suggested one of the solutions from the article at JavaWorld.com, number > 6 or number 7. They allow to control the layout of the jsp page using > xml configuration file, using inheritance when needed. I gave you the > link Sergio, if others would like to take look here it is: > http://www.javaworld.com/javaworld/jw-01-2002/jw-0104-tilestrut.html > As Tiles are still fairly new feature they may be some bugs and > problems, but I guess it's worth trying anyway. > > I'm not sure if that's what You expected Sergio, if not then let me know. > > Regards > Piotr > > > > ------------------------------------------------------- > This sf.net email is sponsored by: Influence the future > of Java(TM) technology. Join the Java Community > Process(SM) (JCP(SM)) program now. > http://ads.sourceforge.net/cgi-bin/redirect.pl?sunm0004en > _______________________________________________ > Edocs-development mailing list > Edo...@li... > https://lists.sourceforge.net/lists/listinfo/edocs-development |
From: pk <pkr...@wp...> - 2002-10-31 11:37:08
|
On 10/29/2002 11:43 PM, Uz.ytkownik Sergio Ramazzina napisa?: > Hi to everybody, > > I appreciate to have a status update on the activities that everybody has > been assigned to. > Can I hope to have something to discuss for the first week of November? On my part: - website is up and running; It has the features You asked for and is up to date. BTW, I have one question - should I register our website with the searchengines? I wasn't sure about that because we're still in early stages of the development so I'm not sure if it should be done now. - user interface - we have agreed on the layout some time ago so I thought we're through with it. About using struts new Tiles feature - I suggested one of the solutions from the article at JavaWorld.com, number 6 or number 7. They allow to control the layout of the jsp page using xml configuration file, using inheritance when needed. I gave you the link Sergio, if others would like to take look here it is: http://www.javaworld.com/javaworld/jw-01-2002/jw-0104-tilestrut.html As Tiles are still fairly new feature they may be some bugs and problems, but I guess it's worth trying anyway. I'm not sure if that's what You expected Sergio, if not then let me know. Regards Piotr |
From: Derek H. <lor...@ms...> - 2002-10-31 09:16:12
|
> I appreciate to have a status update on the activities that everybody has > been assigned to. U9-U15 checked-in. > Can I hope to have something to discuss for the first week of November? How do you want design documentation checked-in? I don't think it should go in the use cases, but rather as a separate programming specification. For example, my programming specification for the Microsoft integ. (MSO and eventually Wexplo) would describe the design of a "package" (I use that term loosely) covering U9-U15. I'm assuming format will follow the use cases, more or less (a big question is what sorts of "headings" you want ... or I can just propose some in my draft). How do we want to structure the source code directories in CVS? I'd like to check the Word prototype and proof-of-concept source code into directories like, edocs/src/proto/interop/mso/ edocs/src/proof/interop/mso/ to keep them seperate from the development code. Prototypes and Proofs are meant to be illustrative, not prescriptive, so they shouldn't be mixed into the Development tree, IMO, since the "production" code may be slightly or wholly different. > Derek, I'm always waiting the requirements for the windows coder you need to > do the hard work. The ideal "Wexplo" programmer, + Expert Microsoft Windows coder. + COM a must. + Windows Shell API a strong plus. Windows Shell API experience will make it easier for the person; approximately 40 hrs coding (I can't estimate the amount of debugging that'll be required tho). A bad shell extension will crash Windows as fast as a bad device driver. Derek Harmon |
From: Sergio R. <sra...@ti...> - 2002-10-29 22:41:27
|
Hi to everybody, I appreciate to have a status update on the activities that everybody has been assigned to. Can I hope to have something to discuss for the first week of November? Derek, I'm always waiting the requirements for the windows coder you need to do the hard work. Thanks Sergio |
From: Sergio R. <sra...@ti...> - 2002-10-28 21:10:31
|
Ok Piotr, Thanks Derek, Ashwini. Any update from you? Fakhar, Narsi, Rahmat, are you able to contact them and start work? I'll wait updates from any of you. Sergio ----- Original Message ----- From: "Piotr Kreglicki" <pkr...@wp...> To: "edocs development" <edo...@li...> Sent: Monday, October 28, 2002 8:24 PM Subject: [Edocs-development] Another website update > Hi all, > > Just wanted to let youknow that I have done some changes to the website > - there is license section now. I have put lists and forums together in > contact us page. > > Regards > Piotr > > > > ------------------------------------------------------- > This sf.net email is sponsored by:ThinkGeek > Welcome to geek heaven. > http://thinkgeek.com/sf > _______________________________________________ > Edocs-development mailing list > Edo...@li... > https://lists.sourceforge.net/lists/listinfo/edocs-development |
From: Piotr K. <pkr...@wp...> - 2002-10-28 19:26:24
|
Hi all, Just wanted to let youknow that I have done some changes to the website - there is license section now. I have put lists and forums together in contact us page. Regards Piotr |
From: Piotr K. <pkr...@wp...> - 2002-10-25 16:54:29
|
Hi all, Just wanted to let you know that I have put gpl license text online in the documentation section. I also got teh news script to work, however I'm not completely happy with it. It will need some extra work to make it work the way I want, but that's just matter of time. Regards Piotr |
From: Piotr K. <pkr...@wp...> - 2002-10-24 09:58:43
|
On 10/24/2002 3:31 AM, Uz.ytkownik Derek Harmon napisa?: >>I think we can use ARGO UML to draw UML diagrams and export files to be >>inserted in the documentation. > > >>Thoughts about that? > > > My first impression of ARGO UML is it's an improvement over DIA > (another open source diagramming package with some support > for UML). Of course, DIA is what made me run out and buy > Visio 2000. DIA was dreadful for UML. ;-) > > While I think it's desirable to have a freely-available, platform- > independent UML solution that everybody can use -- I don't > think one exists yet. There is another tool: Visual Paradigm for UML. It comes in different packages and there is free community edition, it's written in java. The only drawback is it's hungry for resources. Check it out at: http://www.visual-paradigm.com/vpuml.php I display a bit with ArgoUML and liked it because it was light. However I'm not UML guru so I relay on your opinion guys. Regards Piotr |
From: Sergio R. <sra...@ti...> - 2002-10-24 08:43:22
|
Piotr, I would like to have on the project web site a section that we can call "license" with written the complete statement of our GPL statement. You can find it at http://www.opensource.org/licenses/gpl-license.php. Thanks Sergio |
From: Derek H. <lor...@ms...> - 2002-10-24 01:34:04
|
> I think we can use ARGO UML to draw UML diagrams and export files to be > inserted in the documentation. > Thoughts about that? My first impression of ARGO UML is it's an improvement over DIA (another open source diagramming package with some support for UML). Of course, DIA is what made me run out and buy Visio 2000. DIA was dreadful for UML. ;-) While I think it's desirable to have a freely-available, platform- independent UML solution that everybody can use -- I don't think one exists yet. In about five minutes with ARGO UML, I found several bugs: 1. How to name an interface's operation on class diagrams - I enter a name but it keeps reverting back to newOperation. 2. The names of classifier roles keep disappearing, if I switch to a different diagram or connect two classifier roles with a message flow. 3. How to have more than one message flow between classifier roles on a collaboration diagram - the message text appears over previous message text creating mish-mash text).* [*- I played around with it a little more, and although there are no 'handles' on the message flow, if you try and manually contort it out of the way, you can make two message flows visible, so it may not be as much a bug as a quirky feature. ] and several areas where UML support is lacking: 1. No arrow-heads on message flows (to indicate synchronous, flat or asynchronous messages, you have filled, open or one-half arrow-heads) in collaboration diagrams. 2. Can't use the actor pictograph in a collaboration diagram. A workaround is to make the actor a classifier role. 3. Extension use cases aren't depicted on use case diagrams (the split oval notation). Have to add the extension using the extension button, can't just type a new extension use case's name in the "Extension" text box (it just seems to ignore my typing when it has focus). 4. Default values can't be made to appear in class diagrams. From the image file export standpoint, it doesn't support PNG, but GIF, which is in an intellectual property no man's land that most open source projects tend to avoid (Unisys owns the patent on the GIF image compression algorithm. I'm not aware of them enforcing their right to receive royalties from GIF users, but that doesn't mean they never will). The text in the GIF exported diagrams is small and hard to read, there should at least be a way to scale the image that gets exported (but Zooming the workspace to 150% - a quirky technique sometimes necessary in DIA - doesn't improve the exported image resolution). In conclusion, I prefer Visio. I think the ARGO UML tool hasn't yet reached maturity. Derek Harmon |
From: Derek H. <lor...@ms...> - 2002-10-24 01:13:30
|
> I'm really disappointend with someone on how the things are going on. Two > weeks ago I've assigned some tasks to someone of you to take care of > some issues related to the analysis and design of the application. > > The tasks completion date has passed and no one has given an advise on > how his job on the task is going, I'm sorry, I am late with commiting my analysis documentation of U9-U15. I'd still like to apply a GPL notice to the XML files and then check them into CVS under the interoperability tree. Please see my earlier article and let me know what sort of wording is agreeable (re: who or what holds the copyright, and contact information), http://sourceforge.net/mailarchive/forum.php?thread_id=1203596&forum_id=12031 On the matter of designing the MS Office integration, I have begun some experimental prototyping with SOAP from the Word add-in. That design looks like a good one. I'm sorry I haven't checked my e-mail much this week to respond to your earlier questions, but I'm looking into Argo UML now. Derek Harmon |
From: Ashwini K. <ash...@ya...> - 2002-10-24 00:07:14
|
Hello, OODBMS allows you to make java objects directly persistent from Java code. There is no SQL queries involved in the middle. Theoretically, when the complexity of the queries increases or when we find ourselves doing a lot of table joins, OODBMS becomes faster than RDBMS. So, if we find that we have lots of user attributes, permissions and document properties, perhaps OODBMS will be faster than RDBMS. In any case, this is a pure Java solution. Please look at the open source solutions below http://infozone-group.org/ http://www.ozone-db.org/ Ozone OODBMS allows us to store and retrieve XML documents. Also, it implements two phase commit interfaces to support distributed databases. Will let you know soon about ArgoUML. Regards, Ashwini Sergio Ramazzina <sra...@ti...> wrote:Hi Ashwini, I'm really happy to have news from you. I hope you're fine, apart from your job situation. So I usually use Visio for UML diagrams but in these days I've tried ArgoUML, do you know about that?The great thing is that is opensource software, seems good, let you save diagrams in .gif format butthe main thing is that because of it's free we can assume that tool like the official UML tool. Do you agree? Can you try it? You can download it from http://argouml.tigris.org. Have a look and let me know.If also for you could be fine we can use it and I'll take care to rewrite all my diagrams with ArgoUMLand then save them in .gif format. Can you go through the new documentation? We've changed the Word document format to HTML. More lighter and flexible. That HTML is obtained starting from xml files that will be transformedusing an XSL file. Download the updated, compile the documentation with ant and go through. Regarding the database I must say that I'm excited from the idea to use an OO db. Can you giveme an idea about what are the benefits on using that technology in that application respect to using relational DBs? Do we have a strong OO db available to use? Is it possible to have a J2EE applicationalso using OO db? Ashwini, because you're in trouble with your job and you don't have too much time to dedicateto eDocs, of course, I've asked to other three people Rahmat Begum, Fakhar Ahsan and Narasimha Reddy to work on the tasks related to search engine and storage. Can you coordinate them on their job? Rahmat, Fakhar, Narasimha I'll introduce you Ashwini Kumar. He is the manager in charge of the design anddevelopment of the storage and search engine part of eDocs. Please coordinate with him about your work.I hope you do that soon. Regards Sergio ----- Original Message ----- From: Ashwini Kumar To: sra...@ti... Sent: Wednesday, October 23, 2002 10:17 PMSubject: Hello Hello Sergio, Sorry for this late response. I am in the process of moving in between countries I am not in Canada anymore and am in the US. I will soon have to go to India as well to look for jobs. I also have to prepare for interviews and am not very well set up with fast internet access here. But I will try to spend some time for my assignments this week. What tools do you want me to use to build UML diagrams etc? I do not have Visio or any other tool here with me. Perhaps I should just use MS Word to build the diagrams. Are you interested in using object oriented or Relational databases to store documents? Thanks, Ashwini --------------------------------- Do you Yahoo!? Y! Web Hosting - Let the expert host your web site --------------------------------- Do you Yahoo!? Y! Web Hosting - Let the expert host your web site |
From: Sergio R. <sra...@ti...> - 2002-10-23 20:39:21
|
Hi Ashwini, I'm really happy to have news from you. I hope you're fine, apart from = your job situation. So I usually use Visio for UML diagrams but in these days I've tried = ArgoUML, do you know about that? The great thing is that is opensource software, seems good, let you save = diagrams in .gif format but the main thing is that because of it's free we can assume that tool like = the official UML tool. Do=20 you agree? Can you try it? You can download it from http://argouml.tigris.org. Have = a look and let me know. If also for you could be fine we can use it and I'll take care to = rewrite all my diagrams with ArgoUML and then save them in .gif format.=20 Can you go through the new documentation? We've changed the Word = document format=20 to HTML. More lighter and flexible. That HTML is obtained starting from = xml files that will be transformed using an XSL file. Download the updated, compile the documentation with = ant and go through. Regarding the database I must say that I'm excited from the idea to use = an OO db. Can you give me an idea about what are the benefits on using that technology in that = application respect to using=20 relational DBs? Do we have a strong OO db available to use? Is it = possible to have a J2EE application also using OO db? Ashwini, because you're in trouble with your job and you don't have too = much time to dedicate to eDocs, of course, I've asked to other three people Rahmat Begum,=20 Fakhar Ahsan and Narasimha Reddy to work on the tasks related to search = engine and storage. Can you=20 coordinate them on their job?=20 Rahmat, Fakhar, Narasimha I'll introduce you Ashwini Kumar. He is the = manager in charge of the design and development of the storage and search engine part of eDocs. Please = coordinate with him about your work. I hope you do that soon. Regards Sergio ----- Original Message -----=20 From: Ashwini Kumar=20 To: sra...@ti...=20 Sent: Wednesday, October 23, 2002 10:17 PM Subject: Hello Hello Sergio, Sorry for this late response. I am in the process of moving in between = countries I am not in Canada anymore and am in the US. I will soon have = to go to India as well to look for jobs.=20 I also have to prepare for interviews and am not very well set up with = fast internet access here. But I will try to spend some time for my = assignments this week. =20 What tools do you want me to use to build UML diagrams etc? I do not = have Visio or any other tool here with me. Perhaps I should just use MS = Word to build the diagrams. Are you interested in using object oriented or Relational databases to = store documents? Thanks, Ashwini -------------------------------------------------------------------------= ----- Do you Yahoo!? Y! Web Hosting - Let the expert host your web site |
From: Sergio R. <sra...@ti...> - 2002-10-23 08:53:24
|
Dear all, I'm really disappointend with someone on how the things are going on. Two weeks ago I've assigned some tasks to someone of you to take care of some issues related to the analysis and design of the application. The tasks completion date has passed and no one has given an advise on how his job on the task is going, if he has some problems in following the tasks that has been assigned to him (too busy with his job, not much time to dedicate to that, completion date too early), if he has not understood something about the work that has been assigned to him or other. I don't like this kind of behaviour on the project. I appreciate a lot more if the person who has a task assigned informs about the fact that is not able to follow that work for some reasons. This kind of behaviour is more appreciated and honest instead of giving false expectations that no one is able to respect. I think that people works on OSS project because they have fun in sw development. But this doesn't means that we're joking, this is a serious thing and in that way mast be followed. Someone has already been removed from the team and replaced with others. I'm thinking to do the same with some of the rest. Regards Sergio |
From: Sergio R. <sra...@ti...> - 2002-10-22 09:07:55
|
Team, I think we can use ARGO UML to draw UML diagrams and export files to be inserted in the documentation. http://argouml.tigris.org/ Thoughts about that? Bye Sergio |
From: Sergio R. <sra...@ti...> - 2002-10-21 17:07:21
|
Hi all, again I would like to have an update from you Derek and Ashwini about the status of the activities you're involved. Bye Sergio |
From: Sergio R. <sra...@ti...> - 2002-10-18 12:33:08
|
> The GPL recommends the GPL Notice be placed on all source code (ie, > XML source to generate HTML documentation). Refer to the "How to > Apply These [GPL] Terms to your New Program" section in: > > http://www.opensource.org/licenses/gpl-license.php > > What would we want the GPL Notice on source files to say? The > above document recommends the following template, it has a few > blanks to be filled-in: > > - - - > eDocs Document Management System <--name of product?> > Copyright (C) 2002 <name of copyright holder here> > > This program is free software; you can redistribute it and/or modify > it under the terms of the GNU General Public License as published > by the Free Software Foundation; either version 2 of the License, > or (at your option) any later version. > > This program is distributed in the hope that it will be useful, but > WITHOUT ANY WARRANTY; without even the implied warranty > of MERCHANTABILITY or FITNESS FOR A PARTICULAR > PURPOSE. See the GNU General Public License for more > details. > > You should have received a copy of the GNU General Public > License along with this program; if not, write to the Free Software > Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA > 02111-1307 USA > > <how to contact us by paper and electronic mail> > - - - > I know Derek, we'll put that template in every source code we'll write. I'm completing the analysis for the security part of the product. I hope to be able to commit the files in the weekend. In the meanwhile I'm adding the tags you suggested to all documentation files. What's the status for the analysis of the integration part. Are you able to contact Anton? Sergio |
From: Derek H. <lor...@ms...> - 2002-10-18 09:14:49
|
The GPL recommends the GPL Notice be placed on all source code (ie, XML source to generate HTML documentation). Refer to the "How to Apply These [GPL] Terms to your New Program" section in: http://www.opensource.org/licenses/gpl-license.php What would we want the GPL Notice on source files to say? The above document recommends the following template, it has a few blanks to be filled-in: - - - eDocs Document Management System <--name of product?> Copyright (C) 2002 <name of copyright holder here> This program is free software; you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation; either version 2 of the License, or (at your option) any later version. This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details. You should have received a copy of the GNU General Public License along with this program; if not, write to the Free Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA <how to contact us by paper and electronic mail> - - - > > Che ne pensono? > This is the right way to say what you were asking me -> > (Che cosa ne pensi?) > > Ciao Yes. I shall have to confine myself to Ciao, ;-) Derek Harmon |
From: Piotr K. <pkr...@wp...> - 2002-10-17 16:51:48
|
Hi all This is a zip archive with different versions of logos number 2 and 5. I didn't make number 5 in 32px size because "ocs" becomes unreadable, so maybe we should consider just leaving "eD". Regards Piotr |
From: Sergio R. <sra...@ti...> - 2002-10-17 10:21:41
|
> A great logo strikes the viewer when first seen, not read. > > I like logo 4, and also logo 2 without the text. > > A local heart hospital has a logo like 4 (their name starts > with a D), and if it's tilted slightly, it looks heart-shaped. > The other logo looks spiral, like a seashell. > > If we go with one of these, perhaps it can be worked into > a Flash animation and marketing slogans: > > eDocs, the heart of your document management strategy? Piotr, I think we need to consider the ideas of Derek about using Flash and working on our logo. I would like to consider logo 2 to do that (I don't like logo 4). Can you try to do the same work you're doing with logo 5 (preparing two versions one with a dark and one with a light background9? Thanks Sergio |
From: Sergio R. <sra...@ti...> - 2002-10-17 09:21:11
|
> I'd like to recommend 3 minor changes to some of the > use case documents (before we get lots of them ;-) ). > > XML-STYL:ESHEET PROCESSING INSTRUCTION > > Adding a processing instruction will permit Internet Explorer > to automatically apply a stylesheet to an XML document. > > Presently, every use case, taking A1.xml as an example, > starts with, > > - - - A1.xml (excerpt) > <?xml version="1.0"?> > - - - > > I'd recommend adding an xml-stylesheet processing > instruction. This instruction is presently Microsoft-specific > (although it is possible to modify Ant's action to use it > when locating that stylesheet to build docs with). > > - - - A1.xml (excerpt, modified) > <?xml version="1.0" encoding="iso-8859-1" ?> > <?xml-stylesheet type="text/xsl" href="../../../UseCase.xsl" ?> > - - - > > Unrecognized processing instructions don't affect other > applications [XML 1.0 + Namespaces specification], > so this addition should have no impact on non-Microsoft > browsers, while providing Internet Explorer browsers > with added value (IE users can open the XML document > and see the HTML automatically up-to-date). > > While there, an encoding should be added for reasons > detailed in the next section. Derek, XML is just for our internal use.It's the support that we'll use to write raw project documentation and transform applying a uniform stylesheet to HTML documents formatted in a uniform way. The idea isn't that users read directly XML files with thier browsers but that they read HTML. This is valid also for us. That HTML is under the target\doc\ad directory and is obtained after XSL processing with an ant task. In the XSL file I've created some rules that are applied to certain XML element that gives the ability to format HTML in the outputted format (<section>, <detview-section>, <detview-section-item>, etc). Also when we'll have a binary distribution of our product we'll not include the original XML files but just HTML in it. I've followed this method because is the same I'm using with certain Apache project where I'm working on and I think is good. The problems in our case remains again where to live the master files of every diagram in their native format (Visio). I think we can create a separate module on the cvs and live them there. > > > INTERNATONALIZATION > > I'd also recommend adding an encoding (to allow for > translation into other languages and character systems > in the future, for example, Big5 Chinese). Although I > think the Latin 1 (ISO-8859-1) alphabet could be > sufficient for English and Italian. > > Additionally, while this is optional, I'd like to recommend > we start supplying an "xml:lang" attribute with an RFC1766 > lcid to the body element. This way, each body (of which > perhaps there could be several) can be identified with > a language. > > - - - A1.xml (excerpt, with changes) > <?xml version="1.0" encoding="iso-8859-1" ?> > <?xml-stylesheet type="text/xsl" href="../../../UseCase.xsl" ?> > : : > <body xml:lang="en-US" > > : : : > <!-- English --> > : : : > </body> > > <!-- In the future, after use case is stable and translated ... > > <body xml:lang="it-IT" > > : : : > <!-- Italiano --> > : : : > </body> > --> > - - - > > Separate translations (one in English and one in Italian) > are represented as two bodies, one body in English and > the other body in Italian. In XSLT, the lang function can > interpret this, and produce a corresponding HTML META > HTTP-EQUIV "Content-Language" tag. This way, the > browser of the resulting document should see the English > version if the browser's preferred language is English, and > the Italian version if the browser's preferred language is > Italian (I have not tried this out thoroughly). This could be useful when we'll build user manuals or other type of documentation. In any case the language for project documentation is only in english. In any case a good idea is to introduce the xml:lang="en-US" and encoding="iso-8859-1" in the xml. I'll do it immediately. > > > LOCATION-INDEPENDENT IMG ELEMENTS > > The img elements in use case XML documents should not > assume knowledge about where their images are located. > > For example, in A1.xml, > > - - - A1.xml (excerpt) > <detview-section name="Use Case Diagrams"> > <img src="./diags/UC_A1_1.png"/> > </detview-section> > - - - > > Suppose we change the directory where we store .PNG files in > the future? > > All use case XML documents would have to be changed. :-( > > Better that the use case XML document is the image file is > co-located with itself ... > > - - - A1.xml (excerpt, modified) > <detview-section name="Use Case Diagrams"> > <img src="UC_A1_1.png" /> > </detview-section> > - - - > > and we can "hide" our decision to place the diagrams in a diags > folder within the XSLT stylesheet like this, > > - - - UseCase.xsl (imaginary excerpt) > <xsl:template > match='img' > > > <img > src='.\diags\{@src}' > alt='{@alt}' /> > > </xsl:template> > - - - > > What happens now if we change the directory where we store > .PNG files in the future? > > One place, one change -- in the stylesheet. Hmm, what if > there are multiple stylesheets? Your idea is right. My main objective where to leave things related with every single major area separated from the others. So from here I had the idea to have a separate diags directory. If we prefer to have all the diagrams in one single place we need to change the naming convention adding at the beginning of the filename of it's own area name. Ex: security_UC_AI_1.png I like the idea to "hide" the images location and put a reference in the stylesheet to be used in XML. I think we can do that. If there will be the necessity to use more stylesheet file we can build a single xsl file with some "configuration parameters" including it in every stylesheet that we're going to use. I would like to agree on these things asap to terminate very quicly the analysis phase and begin to write some line of code. > > In fact, it could be made a default parameter to the stylesheet, > when we have multiple stylesheets -- and then changes in the > build script percolate everyplace! :-) > > - - - UseCase.xsl (imaginary parameterized excerpt) > <?xml version='1.0' encoding='iso-8859-1' ?> > <xsl:stylesheet > version='1.0' > xmlns:xsl='http://www.w3.org/1999/XSL/Transform' > > > <xsl:param name='diagFolder' >./diags</xsl:param> > : : > <xsl:template > match='img' > > > <img > src='{$diagFolder}/{@src}' > alt='{@alt}' /> > > </xsl:template> > : : > </xsl:stylesheet> > - - - > > Che ne pensono? This is the right way to say what you were asking me -> (Che cosa ne pensi?) Ciao Sergio |
From: Derek H. <lor...@ms...> - 2002-10-17 00:18:16
|
I'd like to recommend 3 minor changes to some of the use case documents (before we get lots of them ;-) ). XML-STYL:ESHEET PROCESSING INSTRUCTION Adding a processing instruction will permit Internet Explorer to automatically apply a stylesheet to an XML document. Presently, every use case, taking A1.xml as an example, starts with, - - - A1.xml (excerpt) <?xml version="1.0"?> - - - I'd recommend adding an xml-stylesheet processing instruction. This instruction is presently Microsoft-specific (although it is possible to modify Ant's action to use it when locating that stylesheet to build docs with). - - - A1.xml (excerpt, modified) <?xml version="1.0" encoding="iso-8859-1" ?> <?xml-stylesheet type="text/xsl" href="../../../UseCase.xsl" ?> - - - Unrecognized processing instructions don't affect other applications [XML 1.0 + Namespaces specification], so this addition should have no impact on non-Microsoft browsers, while providing Internet Explorer browsers with added value (IE users can open the XML document and see the HTML automatically up-to-date). While there, an encoding should be added for reasons detailed in the next section. INTERNATONALIZATION I'd also recommend adding an encoding (to allow for translation into other languages and character systems in the future, for example, Big5 Chinese). Although I think the Latin 1 (ISO-8859-1) alphabet could be sufficient for English and Italian. Additionally, while this is optional, I'd like to recommend we start supplying an "xml:lang" attribute with an RFC1766 lcid to the body element. This way, each body (of which perhaps there could be several) can be identified with a language. - - - A1.xml (excerpt, with changes) <?xml version="1.0" encoding="iso-8859-1" ?> <?xml-stylesheet type="text/xsl" href="../../../UseCase.xsl" ?> : : <body xml:lang="en-US" > : : : <!-- English --> : : : </body> <!-- In the future, after use case is stable and translated ... <body xml:lang="it-IT" > : : : <!-- Italiano --> : : : </body> --> - - - Separate translations (one in English and one in Italian) are represented as two bodies, one body in English and the other body in Italian. In XSLT, the lang function can interpret this, and produce a corresponding HTML META HTTP-EQUIV "Content-Language" tag. This way, the browser of the resulting document should see the English version if the browser's preferred language is English, and the Italian version if the browser's preferred language is Italian (I have not tried this out thoroughly). LOCATION-INDEPENDENT IMG ELEMENTS The img elements in use case XML documents should not assume knowledge about where their images are located. For example, in A1.xml, - - - A1.xml (excerpt) <detview-section name="Use Case Diagrams"> <img src="./diags/UC_A1_1.png"/> </detview-section> - - - Suppose we change the directory where we store .PNG files in the future? All use case XML documents would have to be changed. :-( Better that the use case XML document is the image file is co-located with itself ... - - - A1.xml (excerpt, modified) <detview-section name="Use Case Diagrams"> <img src="UC_A1_1.png" /> </detview-section> - - - and we can "hide" our decision to place the diagrams in a diags folder within the XSLT stylesheet like this, - - - UseCase.xsl (imaginary excerpt) <xsl:template match='img' > <img src='.\diags\{@src}' alt='{@alt}' /> </xsl:template> - - - What happens now if we change the directory where we store .PNG files in the future? One place, one change -- in the stylesheet. Hmm, what if there are multiple stylesheets? In fact, it could be made a default parameter to the stylesheet, when we have multiple stylesheets -- and then changes in the build script percolate everyplace! :-) - - - UseCase.xsl (imaginary parameterized excerpt) <?xml version='1.0' encoding='iso-8859-1' ?> <xsl:stylesheet version='1.0' xmlns:xsl='http://www.w3.org/1999/XSL/Transform' > <xsl:param name='diagFolder' >./diags</xsl:param> : : <xsl:template match='img' > <img src='{$diagFolder}/{@src}' alt='{@alt}' /> </xsl:template> : : </xsl:stylesheet> - - - Che ne pensono? Derek Harmon |
From: Derek H. <lor...@ms...> - 2002-10-16 23:51:48
|
A great logo strikes the viewer when first seen, not read. I like logo 4, and also logo 2 without the text. A local heart hospital has a logo like 4 (their name starts with a D), and if it's tilted slightly, it looks heart-shaped. The other logo looks spiral, like a seashell. If we go with one of these, perhaps it can be worked into a Flash animation and marketing slogans: eDocs, the heart of your document management strategy? Derek Harmon |
From: Sergio R. <sra...@ti...> - 2002-10-16 13:55:02
|
Team, based on some Derek's observations about the way we can build project documentation I've decided to radically change the way project's documentation is written now and for the future. The way to write documentation is to use XML with an appropriate XSL file and an ANT task to process the builds of a series of HTML files that we and other people can read with a simple browser. This approach gives us: 1) Flexibility in writing documentation without platform specific issues 2) Less space required 3) Uniform project wide docs layout given by a common XSL file and a common stylesheet. First of all the documentation directory's name has changed from "documentation" to "doc". I've asked Sourceforge team to delete the oldest one. The new directory has a subdir called ad that is the root of the Analysis & Design documentation. There we have an index file that links all the 6 application areas: storage architecture, search engine, security, system setup, interoperability, web interface. Under there we have one subdirectory for each one of the five areas. doc | ---- ad | ----- storage | ----- search | ----- setup | ----- security | ----- interoperability | ----- web_interface Every subdirectory contains core A&D documentation for that specific single area. Every subdirectory is organized with an index file and a "uc" subdir where we have use cases for that specific area. The name of the use case documentation file is like the identification number of the specific use case (ex. for A1. Add new user the uc file will be named A1.xml and so on). Under the "uc" subdir we'll have a "diags" subdir that contains all the UML diagrams needed for the specific application area (use cases, sequence, collaboration and so on). All the diagrams will be in PNG format. The naming convention for every diagram will be <diag_type>_<use case id>_<progressive>.png So for example the first use case diagram for A1 will be named UC_A1_1.png. I suggest for <diag_type> to use UC for Use cases, COLL for collaboration diags, SEQ for sequence diags, ACT for activity, CL for class diags, DOM for domain diagram. So if we consider the use case A1. Add new user it will be described in the file A1.xml contained in doc/ad/security/uc. The use case contains a use case diagram that will be named UC_A1_1.png contained in doc/ad/security/uc. The processing of the xml files to obtain HTML will be done using the ant task compile.manuals. The resulting HTML files will be put under the path target/doc/ad and so on (same struct as above). The XML files will be composed using some simple xml elements. You can take a look at the files already present to have some example on how to do that. The use case A1 really contains diagrams. So take a look. The original word document has been eliminated and all the UCs has been migrated in this new format (also the Derek's ones). Derek what do yuo think about that? Also I would like to have feedbek on the tasks I've assigned. Derek, Ashwini, Anton, Dobb can you give me news?? Piotr, can you take a look at the files and give proposal for a much better layout. It maust be simple, without graphics (just our logo, but much better navigations mechanisms. Regards Sergio |