Reference Implementation Meeting 2011-02-01
Jump to navigation
Jump to search
RI Workgroup Call
Tuesday, February 01, 2011
12:00 PM EST
Attendees: Will Hartung (Mirth Corp), Sri Koka (Techsant Technologies), Mark Bamberg (MEDfx), Patrick Pyette (Inpriva), Greg Meyer (Cerner), Brian Hoffman (DoD), Beau Grantham (Harris), Umesh Madan (Microsoft Corp), Chris Harris (NextGen)
Beau -
Not a whole lot going on since last week.
Umesh is busy with go-live for HealthVault. Not a lot of progress on .NET side.
For Java side - Greg has been checking things in around the file auditing and MBeans.
Greg -
There is now an auditing interface in the Java implementation. It's gone through a second enhancement. There is an implementation of file base one that uses the default logger, another uses a more sophisticated logger, and a no off auditor used for unit testing. Again, those weren't really intended to go into a production environment, but if anyone wants to write an ATNA implementation you're more than welcome to. It's like anything else on the Java side. I extended it to use Java SPI.
We have a fix in for the date field in the MDM. That's only in the 1.2 release. I'm looking at moving it back to a 1.1.2. I'm going to bring up the status in terms of releasing and what the strategy is. A lot of the code that's in there now is not going to be in the 1.0 release, so we'll talk about how that's going to go. Other than that, I don't really have any other updates. I'm working on Cerner Direct, our production HISP.
Beau -
Last week on the Java call we talked to Tony at the VA and they've had success implementing Java implementation in their product as well, so good progress on their end. We also have Vince back for the Java team. He'll be helping us out to XD. That will be the key to getting that to production.
We have a meeting after this to discuss the XD status to see what needs to be done, where we need to go, etc. so we can get that out ASAP.
Are there other general status
Sri-
We're cleaning up the documentation. We're working to see how to use it, how to interface with the HISP, EHRs, etc.
Mark -
My main goal this week is to get ready for Virtual connectathon so we're ready to interact with other people. I've been focusing more on our demo for HIMSS this week.
Beau -
REMINDER: We've got a virtual connectathon this Friday. We'll hopefully get through the rest of the negative test case scenarios; we'll go through the full suite. An outcome of that will be any bug fixes necessary for a 1.0 release which is scheduled for Wednesday of next week to give us a few days to get all of the bug fixes in. We don't have anything outstanding -- it'll just be the few things that come out of that. Once we get the status on XD, we should quickly follow with a release of the XD components. Other high-level topics they want to discuss?
Greg -
In terms of release and the code that's sitting in material vs. what's going out in 1.0, we've pretty much have all of the components in place for the 1.0. In terms of new code for 1.0, there's nothing new that we're checking in barring any bug fixes coming up this week. What we're looking at doing is taking the M2 release from the Google Code repository and taking out the XD components. What does that mean in terms of the code material? I've seen a lot of folks checking out the code and doing a clean install. The code that's currently in material is stuff that we're targeting for future releases and the 1.1 release. 1.2 will not be on the 1.0 stock assembly release so for those of you that are pulling code out of material, that's not 1.0. There is another component. The new Direct Comments, we haven't had a chance to update that project yet with the latest components. If you are trying to run that you may run into a few issues. There are a couple new dependencies, specifically around auditing. The only item that I do know that will probably go in 1.0 is the date fix. So that's the high-level release strategy.
Beau -
So the auditing and things that have gone in recently are for a future release?
Greg -
Yes. There hasn't been enough testing for me to feel comfortable putting that on the release next week.
Beau -
I know we also discussed last week about some performance metrics and things like that. Any updates?
Greg -
No, not yet. There have been some questions about hardware requirements for different installations. We don't have any data to push back to anybody. I'm not sure if this is something we should or should not be undertaking on an open source project like this.
Beau -
Regarding that, I think we responded to others with the minimum set of requirements that we know we've tested on so they have a baseline of what can run it. We pointed them to deployment models and things like that where they would place it in their existing enterprise and basically go from there, but we didn't give them a fixed set of what you need because we just don't have that at the moment.
Mark -
I think I documented something else, though I'm not sure if I specified what the hardware OS was. I have a general question. Right now I'm looking at the code trying to integrate to my HISP. Where is the place to put a hook so you can inspect incoming emails after they're received and decoded? I want a hook to call our web services when e-mails arrive.
Beau -
That would be done inside the James configuration...
…
Beau -
Umesh, can we get an update on .NET
Umesh -
We've had some check-ins, bug fixes. I'm basically slammed on internal Microsoft stuff. So that's pretty much it from me. The odds are high that I will not be at the connectathon. I'll have to catch up with folks independently. Our HISPs will be up. The main thing is that I think all of the anchors from last time are still there. Hopefully, unless there were new ones checked in recently, they should all be there.
Greg -
I won't be able to be their either. Our HISP will still be up.
Umesh -
I have a couple of improvements. I will check them in sometime next week. I've generalized the routing mechanism… so it should be more resilient to downtime.
For XD call: I do have BLOB source…
Beau -
The purpose of that meeting was to see where we are in status and what we can do to get the ball rolling.
Umesh -
We're having a resource crunch here. What I would say is what we really need help with specifically is: 1) if someone can help with documentation 2) ...
Beau -
We do have the plan for the 1.0 release Wednesday of next week.
Umesh -
That's where we are.
Beau -
That's all I had as well. Anything else?
Umesh -
The week after next will be better for me. I've hired a new developer, so we'll see when he can start.
Beau -
I think we're in good shape for a 1.0 release next week and we'll try to get the ball rolling on XD after this meeting. If anybody wants to join that it's the top of the hour on the same number.
Tuesday, February 01, 2011
12:00 PM EST
Attendees: Will Hartung (Mirth Corp), Sri Koka (Techsant Technologies), Mark Bamberg (MEDfx), Patrick Pyette (Inpriva), Greg Meyer (Cerner), Brian Hoffman (DoD), Beau Grantham (Harris), Umesh Madan (Microsoft Corp), Chris Harris (NextGen)
Beau -
Not a whole lot going on since last week.
Umesh is busy with go-live for HealthVault. Not a lot of progress on .NET side.
For Java side - Greg has been checking things in around the file auditing and MBeans.
Greg -
There is now an auditing interface in the Java implementation. It's gone through a second enhancement. There is an implementation of file base one that uses the default logger, another uses a more sophisticated logger, and a no off auditor used for unit testing. Again, those weren't really intended to go into a production environment, but if anyone wants to write an ATNA implementation you're more than welcome to. It's like anything else on the Java side. I extended it to use Java SPI.
We have a fix in for the date field in the MDM. That's only in the 1.2 release. I'm looking at moving it back to a 1.1.2. I'm going to bring up the status in terms of releasing and what the strategy is. A lot of the code that's in there now is not going to be in the 1.0 release, so we'll talk about how that's going to go. Other than that, I don't really have any other updates. I'm working on Cerner Direct, our production HISP.
Beau -
Last week on the Java call we talked to Tony at the VA and they've had success implementing Java implementation in their product as well, so good progress on their end. We also have Vince back for the Java team. He'll be helping us out to XD. That will be the key to getting that to production.
We have a meeting after this to discuss the XD status to see what needs to be done, where we need to go, etc. so we can get that out ASAP.
Are there other general status
Sri-
We're cleaning up the documentation. We're working to see how to use it, how to interface with the HISP, EHRs, etc.
Mark -
My main goal this week is to get ready for Virtual connectathon so we're ready to interact with other people. I've been focusing more on our demo for HIMSS this week.
Beau -
REMINDER: We've got a virtual connectathon this Friday. We'll hopefully get through the rest of the negative test case scenarios; we'll go through the full suite. An outcome of that will be any bug fixes necessary for a 1.0 release which is scheduled for Wednesday of next week to give us a few days to get all of the bug fixes in. We don't have anything outstanding -- it'll just be the few things that come out of that. Once we get the status on XD, we should quickly follow with a release of the XD components. Other high-level topics they want to discuss?
Greg -
In terms of release and the code that's sitting in material vs. what's going out in 1.0, we've pretty much have all of the components in place for the 1.0. In terms of new code for 1.0, there's nothing new that we're checking in barring any bug fixes coming up this week. What we're looking at doing is taking the M2 release from the Google Code repository and taking out the XD components. What does that mean in terms of the code material? I've seen a lot of folks checking out the code and doing a clean install. The code that's currently in material is stuff that we're targeting for future releases and the 1.1 release. 1.2 will not be on the 1.0 stock assembly release so for those of you that are pulling code out of material, that's not 1.0. There is another component. The new Direct Comments, we haven't had a chance to update that project yet with the latest components. If you are trying to run that you may run into a few issues. There are a couple new dependencies, specifically around auditing. The only item that I do know that will probably go in 1.0 is the date fix. So that's the high-level release strategy.
Beau -
So the auditing and things that have gone in recently are for a future release?
Greg -
Yes. There hasn't been enough testing for me to feel comfortable putting that on the release next week.
Beau -
I know we also discussed last week about some performance metrics and things like that. Any updates?
Greg -
No, not yet. There have been some questions about hardware requirements for different installations. We don't have any data to push back to anybody. I'm not sure if this is something we should or should not be undertaking on an open source project like this.
Beau -
Regarding that, I think we responded to others with the minimum set of requirements that we know we've tested on so they have a baseline of what can run it. We pointed them to deployment models and things like that where they would place it in their existing enterprise and basically go from there, but we didn't give them a fixed set of what you need because we just don't have that at the moment.
Mark -
I think I documented something else, though I'm not sure if I specified what the hardware OS was. I have a general question. Right now I'm looking at the code trying to integrate to my HISP. Where is the place to put a hook so you can inspect incoming emails after they're received and decoded? I want a hook to call our web services when e-mails arrive.
Beau -
That would be done inside the James configuration...
…
Beau -
Umesh, can we get an update on .NET
Umesh -
We've had some check-ins, bug fixes. I'm basically slammed on internal Microsoft stuff. So that's pretty much it from me. The odds are high that I will not be at the connectathon. I'll have to catch up with folks independently. Our HISPs will be up. The main thing is that I think all of the anchors from last time are still there. Hopefully, unless there were new ones checked in recently, they should all be there.
Greg -
I won't be able to be their either. Our HISP will still be up.
Umesh -
I have a couple of improvements. I will check them in sometime next week. I've generalized the routing mechanism… so it should be more resilient to downtime.
For XD call: I do have BLOB source…
Beau -
The purpose of that meeting was to see where we are in status and what we can do to get the ball rolling.
Umesh -
We're having a resource crunch here. What I would say is what we really need help with specifically is: 1) if someone can help with documentation 2) ...
Beau -
We do have the plan for the 1.0 release Wednesday of next week.
Umesh -
That's where we are.
Beau -
That's all I had as well. Anything else?
Umesh -
The week after next will be better for me. I've hired a new developer, so we'll see when he can start.
Beau -
I think we're in good shape for a 1.0 release next week and we'll try to get the ball rolling on XD after this meeting. If anybody wants to join that it's the top of the hour on the same number.