I have been doing some thinking on this freebay idea and came up with some things. As I said I am not against working on a complete application, however I think we should focus on an app that has potential resale value. Why? Simple.... if it can be sold developers arent working for free, and it brings something back to jboss to keep the project going. In this vein I came up with a couple ideas. Both focus on the public sector. (I knew those 16 years I spent there would pay off sooner or later [:)]
The premise is simple... here in the US (and probably everywhere) government spends millions on custom development. However here in the US we have a law that if a commercially available ots solutions exists that fits a need (and assuming it is less expensive) they have to purchase the ots rather than custom develop. So I have 2 ideas... 1 I know is very viable... 2 I have not researched deeply.
idea 1) a grant submission application. Every agency in the US government has some form of grant or loan that they administer. some agencies administer billions every year. Until recently this has all had to occur offline. Now the government has to start accepting grants online. I know of only 2 agencies that currently accept grants online The remainder (as far as I know) are only using fillable PDF which can be emailed. So there is a tremendous opportunity for someone to come in with a solution here.
idea 2) Since I live in the DC area, and spent a great majority of my adult life in the government sector I hear a lot of rumblings. One that is starting to get really big is HIIPA compliance. Now HIIPA is some health thing. I am not quite sure what kind of app would fit in this space since I havent researched it deeply, but there are an awful lot of rumblings going around about needing something to assist in this.
My proposal is simple... instead of doing freebay (which sounds like a cool app by the way) why dont we consider something that has resale value, and hence the potential to earn us all a little in the long run.
I have considered writing app 1 myself. Since I have been involved in one effort to write and deploy such an app i have the domain knowledge. (actually the team i led wrote the app that is used at the department of justice for this purpose http://grants.ojp.usdoj.gov - granted most of it is hidden behind a user login, but you can see the basics there on help pages) So I can come up with a quick list of requirements. Plus this one has international possibilities. Every country has some kind of grant process. To my knowledge there is only one other company playing in this area. A place out of Canada. Their solution is not directed at grant management specifically, it is more of a "we can make it fit that model" kind of app. I believe I can find us the vehicle to sell such an app. (after all I live here in the heart of beauracracy [:)]
Now dont get me wrong... this is still a free kind of app. People could see all the source and such, but for commercial use they would buy it.
If there is any interest in this kind of vein let me know.... This is not a complex app believe it or not and could probably be written in a couple months max. (Heck we wrote the one at DOJ in a month with 10 people) potentially here we have more than that.
Now... if I am totally loony feel free to let me know. But I really think this would work. Plus it gets jboss into use in some really viewable places. The only thing i ask... assuming noone thinks I am loony, is please dont bother if you dont want to help out. Im with Marc on code talks and BS walks.
In that vein Im still working on My first "demo" stuff for Jboss. (its harder since simultaneously I am writing an article for Javaworld on the same topic) And hope to have that ready tomorrow.
So... any takers on this idea?
If we write an example app, then it should have some kind of usefulness otherwise it wouldn't be an example app, code snippet are best written by the original developers or by the ones who want to learn the code.
I want to participate in jboss but there must be a design plan, this may be difficult in opensource but isn't impossible.
I think that the best use of an example app is how an application comes to live, why some design decision's are made, how that plan is turned into real code. I personally think that there got to be a raison after each decision not because there is a good design pattern that say so. I call that "common sense pattern's".
Do not misunderstand me, design pattern do have there value, if you don't forget the original idea of the pattern.(I look at design patterns as examples).
the real value here is that it must be educational.
No probe can fail, because there is a raison (even bat ones) why action toke place.
> If we write an example app, then it should have some
> kind of usefulness otherwise it wouldn't be an
> example app, code snippet are best written by the
I agree completely. The proposal is not just for a sample app. It is for something that has commercial value. This is an app which could be resold.
The code samples are also being worked on. Actually I just uploaded one to the jboss/patches. A simple message driven bean example and web test client.
However my proposal here was for a full blown app. It is open source, but not free. there is a big distinction here. :) the app would be open source, and we would have to come up with some sort of license. however it would be for resalem and the copyright and certain rights retained. (the developers of course getting proceeds, this is the model Marc is using on the documentation. I think it is a good solid model.)
Just because something is developed and the source is available doesnt mean you cant charge for it :)
The problem here is, who wall earn the profits?
this is a harder question than it seems.
because if you sell a original free product what about the people who have contributed for free?
I find your message interesting but for different reasons than you may have intended. It was your mention of HIPAA (Health Insurance Portability and Accountability Act) that really made me take notice.
I am working a project right now that takes on this HIPAA business as one of its requirements. We are building this application on JBoss for a federal client.
You mention a desire to get JBoss "into use in some really viewable places". Even though I am a JBoss user and not part of the development team, I share your desire because it helps me.
I think I can understand why you want to do this, but I would like to hear it in your words. Will you expand?
I'm interested in working on the "grant submission" app, it sounds like a good idea that is small enough to finish quickly. I'm still learning JBoss, but I know Java and some J2EE.
I missed the FreeBay discussion, anyone have a link to it?
Also, can you expand on:
>Just because something is developed and the source is >available doesnt mean you cant charge for it
I can find myself in the fact public software must kept public. in the sense that every citizen in a democratic nation must be able to track down what the government or better government companies do with there personal data. I known that different nations are looking in that direction.
Where i live (europe) they are talking about that but there aren't any standards yet. So let follow the standards that us government has published.(with the assumption that there may be other standards included lathers)
Personally i not that keen in government administration
but there is certainly a future in it.
which options do we have?
are there any legal condition to known about?
opensource for the public!
first note on the todo list: research
grant submission app, that's what i was into for my last contract. my experience:
- the client preferred something off-the-shelf than one built from scratch. i searched all over the place. the closest i could find were paper submission systems (for conferences) and proposal submission systems which didn't quite fit the requirements. and either these apps were in prototype stage or just too expensive. and there was one in the U.S. but it was for internal use only. i think there are lots out there but hiding in intranets.
- indeed, a grant application submission system is not that complicated. i and another developer wrote two versions in one summer (and in Canada, summers are short ;). i would have liked to do the review system too but the client still preferred to do that over meetings, at least in the beginning. and it seems that they were comfortable with applications coming in but not going out.
- security and privacy of high importance with this app, what with research proposals, business plans, resumes, reviews, $Ms involved. not just anybody off the street can register and submit. you must have some blessing and authorization first before you can submit. the system has to be in-house and mostly internal. no way they're gonna use a portal-type web service. that is, this grant submission app is not gonna be very visible, at least with this particular funding group. so i'm not sure how well this app will do for marketing purposes.
i don't have much jboss experience but i can probably provide some Canadian flavor to the domain knowledge if you think that will help. i'm interested in this application and i'm sure my client will be too. you can at least be sure that i will try make them get a commercial copy if/when one becomes available.
I would be willing to help on a grant application. As long as we go with the aim of incorporating as many JBoss features and the *correct* use of them as we can. I want to learn more J2EE/JBoss and the application we do doesn't really concern me as long as it's achievable.
Interesting, I am willing to contribute to this if it can be done. But there has been posted some questions around this that need to be answered first.
My personal view would be that this should be freeware and follow the same principals as JBoss dose. Perhaps this "application" can be developed as a subproject to JBoss 3.0?
Can anyone give a grant submission senario ?
I would like fully understand what such applications
Ok First off let me appologize. I have been out of pocket for a bit due to a contract being canceled and having to finish up early what the client wanted. Now that I have a breather I will try to summarize all the answers in this one posting :)
Working backwards, and from memory here so if I miss something just ask again, I will try to review the posts to make sure I hit everything but invariably I will miss something.
1) Why do I want to see Jboss in visible places? Well its twofold. I have been following the project for some time. I have used it as a dev server in several projects, but couldn't ever get the client to bite on it. (sometimes I dont know why the only thing it was missing was clustering, and 90% of the clients i have worked for dont even bother with clustering) So it helps JBoss for one. And with few knowledgeable JBoss developers around (in the DC metro... if you are there drop me a line :) one can never know too many people.) so it helps me in the long run.
2) What is a grant submission app? Well this is a complicated question, since we are dealing internationally now. At least we have some experience in Canada so that will help us. Here in the US you have several forms to submit. SF-424 Series. The exact forms depend on what type of grant (construction or non-construction). Now there are several important points here. Once a grant application has been assigned an application number the government has the requirement to track it. Whether money is awarded or not. So a simple timeline would be...
Fill out Sf424, sf424-x (where x is the A or B etc)
Pending application goes into a review bin
review application (government agent)
deny application (not qualified etc) or approve as legit and assign app number.
On the submission side that is it. Now the review side I have experience in also. but Lets see if we can get a submission app going and then look at the review.
It is NOT a complicated application. What will make it the most complicated is allowing pluggable forms for different countries. So step one would be, as someone suggested research. There is a very exhaustive data dictionary here in the US that describes all possible fields of a grant application and grant. So that is not too bad.
3) HIPPA. - Have fun :) Actually I would love to get on a HIPPA project. Or contract management project. (Similar really to grant management) Is this a state government or federal? Just out of curiosity.
4) on the public source but charging. Its pretty simple. Open source in general means open source. Since this is an off the shelf app, primarily intended for the government it is quasi open source. Someone (imho preferably jboss group) will retain rights to the source code. Therefore although it is open source, and available you cannot use it in a production environment without paying. Lets face it, Governmetn pays hundreds of thousands to develop these things custom (I actually had a $3.2 mil contract on the last Grants management contract I worked on... before budget crunch hit and it was canceled. :) Developers can still see how to do things in Jboss, Jboss gets a packaged App that can be used to market Jboss, and there is potentially some renumeration for developers. The question came up about how to renumerate... well i havent worked out the details. I had a brief conversation with Marc about the idea, but we havent developed it fully. In short it would NOT be difficult to track who is working on a small project like this. So I doubt it would be difficult to figure out something fair. And frankly there is no guarantee there will ever be renumeration. So it might be moot.
5) a subproject of Jboss - well yes and no. I would target Jboss 3, simply because that will be the latest release. But I wouldnt tie it into the JBoss licensing because of the above listed renumeration reasons. (Im not against free software at all, in fact i love it :) The potential of renumeration will possibly attract some decent coders too though.
6) So... i think - in looking over the posts that covers everything.
Meg - Yes I think the canadian domain knowledge would be great. it will definitely help out. As you pointed out this is not a difficult app. Last year we put one together in a month with 5 developers.
OK, I came up with an initial very quick set of requirements. this is off the top of my head using the domain knowledge i have. It is probably not exhaustive :)
1) User will be able to log on using their user id and password
a) If user has no user id they should be able to create an account.
b) Accounts can be either automatically activated or need approval by use of a configuration option
c) User can be either a point of contact or the signing authority.
2) Once user is logged in they should see all their pending applications and status
3) Applicant can view funding opportunities and apply for grants under them.
a) Forms used to apply for application will be pluggable
b) Application should accept file attachments (Some agencies may want budget information uploaded instead of interactively filled in)
c) Each application will be associated with the user, and either a point of contact (If the user is the signing authority) or a signing authority.
d) User will be able to place an uncompleted application in a holding area in case not all information is available when filing.
e) User will be able to review completed application prior to submission.
f) Upon submission application will either be automatically assigned an application ID or placed in a holding bin.
1) If placed in a holding bin application is assigned an application id upon approval.
2) Application ID's are a sequence of alphanumeric characters, the pattern of which can be set by a configuration parameter.
4) Funding opportunities have distinct start and end dates for application submission. Applications are not accepted outside these times
5) Funding opportunites can have free form text or one or more downloadable items associated with them.
6) Users come in several roles -
a) Applicants - people who apply for grants
b) Grant managers - people who can approve applications and user ID's in thier program areas
c) Administrators - people who can post new funding opportunities
These all come from things we either did or later the client came up with as a need. So These are basically real world.
>HIPPA. - Have fun Actually I would love to get on a >HIPPA project. Or contract management project.
>(Similar really to grant management) Is this a state >government or federal? Just out of curiosity.
It is federal. One of the departments has an archipelago of clinics stringing across the country. We have been funded to build an e-Health application for them.
The approach we are taking involves implementing the ANSI standard object model for healthcare, which to date, I don't believe anyone has done and certainly no one has done using J2EE.
My vision, is to have an implementation of this standard available as OSS. That is everything from the OS to the App can be had for free.
In January, I'll be going to a conference where I'll be seeking others who are interested in seeing this come about. At the last conference, where I was on a much smaller soapbox, I still managed to light the wires of Sun microsystems and the representative from the Univ. of Warsaw. So I am cautiously optimistic about this time around.
I think our efforts here are giving JBoss some part its deserved visibility. Certainly, the federal mucky-mucks are hearing the virtues of OSS, which has loomed large in my last two presentations.
By the way, the acronym is HIPAA (Health Insurance Portability and Accountability Act).
If your cause is to help jboss get exposure i think any help is good help. Currently I am working on a project using JBoss and Tomcat. I started it initially to get hold of things and seeing everything in the perspective of a J2ee architect.
The site has about 40 distinct pages with discussion groups and messaging to other users. I cannot reveal all the details untill it is released but hey - i will post it here when it is done. I am currenty working by myself and i am not expecting any help.
Hope to finish soon and post the URL .