Hi Reza, I shall take this up. To another point that Werner mentioned. I can also pitch in to add openDDR support to module that you have written.
Regards Gopal From: Reza [mailto:[email protected]] Sent: Saturday, April 05, 2014 8:59 PM To: gopal Kalyanaraman; Apache Device Map DEV Subject: Re: Better documentation. Yes, I would love too :) So there are 4 use cases for the website: 1) As Venkata mentioned, information. Examples, downloads, roadmaps. Nothing special here, just content. 2) Users use the website to test user-agents and classify them. This could even be a public service. They input a user-agent, we tell them what the device is. This service will use our Java client and the latest DDR data. 3) If the classification in step #2 is wrong or we simple dont have the device in the DDR data, the user can then flag the user-agent for "developer review". This submits the user-agent to a database for a developer to look at. 4) Developers use the website to review user agents. A developer will either fix the broken pattern or add a new pattern. This process will drive the DDR data releases! So I attached a very simple Java/Spring/Hibernate framework I wrote last year to kick off dev of the website. It runs in Tomcat and just needs a database behind it (I was using mysql, but anything hibernate supports will work). It just sets up a basic database driven MVC app. Im certain once this gets a bit of momentum, things will start to pick up. I definitely see a use for this project, we just need to get to the 80% done mark. Right now I think we are around 40%... ________________________________ From: gopal Kalyanaraman <[email protected]<mailto:[email protected]>> To: Reza <[email protected]<mailto:[email protected]>> Sent: Saturday, April 5, 2014 10:58 AM Subject: RE: Better documentation. Hi Reza, Can you give me some more info on the website that you are expecting to create. I can pitch in. Regards Gopal -----Original Message----- From: Reza [mailto:[email protected]<mailto:[email protected]>] Sent: Saturday, April 05, 2014 7:42 AM To: [email protected]<mailto:[email protected]> Subject: Re: Better documentation. hey Venkata, thanks for reaching out. I understand your points and I agree that the public facing aspect of this project needs some work. So I think we need to address some of other aspects here which is causing the lack of focus on things like the website. Myself and Eberhard spent some time last year rewriting the core clients. While that effort was successful, we kind of got bogged down in some of the data correctness of the clients. Its not that the new clients aren't correct, they are just different than the legacy OpenDDR clients. So they aren't a drop in replacement, rather, its a rewrite. In my opinion, everything legacy should probably be discarded code wise. I can easily do this, but see my next point... We also have the issue of the actual DDR data. The OpenDDR data is good, my dClass project easily supports it and gets very high levels of accuracy. But I think it would be good to get that process moved over to this project. A big part of doing that is creating a website where we can have users submit new devices and their user agents and then have developers review them and add them to the DDR data. Having the clients and data under one set of developers would be ideal for this project. So I think what you are asking for falls into my 2nd point, the website. At this point, I only have bandwidth to maintain the client. I think we need someone to champion this website which will really kickstart this project. Are you interested in taking a shot at this? The key here is the website needs to drive how we maintain the DDR data. Obviously once we get it up and running, myself and others can fill in the details like roadmap, examples, etc. thanks, Reza ________________________________ From: Venkata Kiran <[email protected]<mailto:[email protected]>> To: [email protected]<mailto:[email protected]> Sent: Friday, April 4, 2014 3:22 PM Subject: Better documentation. Hi, First I would like to thank all of you for this initative and contributing to this interesting project. But I think a few steps are required to make this reach for more developers. The purpose of the project is not very clear. It is stated in 2-3 lines, but I don't think it is easy to understand unless some one knows about similar requirement or library before. I think it should be more descriptive on what this project can do. Getting started guide: There is no good Getting started guide. So who ever wants to use this library there is not enough documentation to get them going. This is available in couple of steps but I don't think new developers hoping to use this will feel comfortable with this information. They may require more descriptive instructions to start using this. The easier the developers can start using the library, the more the number of developers will be using. The more the number of developers using the more the feedback the committers can receive and the library will mature with the number of users. RoadMap: The road map of the project is not clear. I think this is required to give a clear direction to the potential contributors, which has the ability to rope in more contibutors. I am sure these are on your list of things to-do, but I think this require more attention. The current documentation on the website doesn't give any comfort to use this framework even though the framework itself is good. I understand it is little difficult to keep the documentation up to date when the project is evolving fast, but I think making this happen will help several developers who want to try this out. I don't want to be a critic on this, but these are the problems I faced while starting on this. I would love to help on these , but I think these have to be done by the committers working on this for long enough so that the intial drafts can be created correctly. Once these are in place, I believe there will be several other developers who can step in to contribute. Thanks, --Kiran
