
07:36:28
21st & 22nd September for the next hackathon on Identity

07:44:46
Helen Can you post the Hack twitter handle-thanks

07:50:59
Twitter Handle #FHIRHack2020

07:55:37
Does this rely on the EPMA system defining AMP?

07:56:05
Bye all-Thanks for contributi

08:03:25
I'm not going to be able to present or create slides as I need to run but I have got a video of the result.https://www.awesomescreenshot.com/video/407847?key=3a63964879895ccd164f4a3bfa9a163aUsing NHS login then pulling down meds from the orion consolidated record and presenting to patient as a list that can be edited (mark medicines as inactive / add comment).

08:04:08
Thanks Matt for being part of the event and leaving your video

08:14:22
Thanks Matt, great to see our new API being used!

08:27:14
How long do messages stay on a queues?

08:27:20
^Kafka

08:38:41
@kevin

08:38:41
It is configurable, - by default 7 days

08:38:54
Or infinite - it’s a log rather than a queue

08:39:38
we are thinking along the lines of Kafka topics being the longitudinal record

08:39:55
Thanks. The Manchester LHCRE FHIR Server has similar topology. We don't use an external queuing engine at the mo but planning too.

08:41:09
One of the benefits of kafka is the connector ecosystem so you can push data into and out of kafka into the tools that best suit the proble you are trying to solve

08:41:32
What happens after the end of that time period in Kafka? Does the data get persisted anywhere?

08:42:29
You have different options - so you can persist it for long lengths of time - moving it into cheaper tiered storage and/or you can move it into different data stores or both

08:43:13
Its for queuing/streaming though, not for persistence.

08:43:46
The multiple consumers may be very useful.

08:44:52
Apologies everyone but I am going to have to leave the demo. I will video the presentation and share. Some really great stuff happening here!

08:46:04
Hi RobI've just seen your slides and added you to the list. There will be time if you want to share your hack. We're just running a little tight on time but sure we'll be okay.

08:46:23
One of the things you need to consider is if you need to preserve the historic state of the data in order to rebuild information - this can be useful if you want to recreate datastores after schema changes as you can rebuild the database rather then mutate (update it). Generally Kev is right you don't look to hold forever for most use cases

08:47:41
At the top of the call View Options - Select Mark Hunt

08:48:15
Mike, what we have at the mo is on this url.basically it's Apache Camel+HAPI going via queue to HAPI JPA FHIR Server.Having kafka in between the two would probably work

08:48:19
https://project-wildfyre.github.io/careconnect-messaging-r4/

08:48:49
Camel is doing the transforms

08:49:40
Thanks Kevin that's a perfect paring!

08:51:15
One of the things we think will be helpful is to use CADE to help model future APIs quickly or to model non FHIR flows as this will help us understand how to handle a mixed economy of interoperability standards

08:53:34
Is CADE open source?

08:57:36
Thank you everyone! This has been so educational (and a little bit fun). But I have a kid who needs picking up!

08:58:09
Hi Steve - the CADE engine is not open source, but the configuration files, pathways and generated outputs are…

08:58:26
:)

08:59:04
the CADE engine is (and will remain) available for free use as long as the config files are kept open

09:09:32
Brilliant!

09:18:18
I think Better use Kafka from memory. Would be good to confirm

09:21:05
I wish I had Better.care available as a test environment! Looks awesome

09:22:03
Sorry I mean OpenEP

09:23:45
It is really good set of products to work with and a great team too!

09:26:17
yep

09:28:56
:D no problem

09:33:17
Thank you though, I appreciate it!

09:33:42
I wanted to give David Beaumont a vote too :)

09:33:56
As did I

09:34:17
David Beaumont you would have got my vote also. Great to see

09:34:45
Ditto, David Beaumont you would have got my vote too. :)

09:35:56
Is this where we all network and drink free wine at the end :) ?

09:37:13
Thanks very much for the vote of confidence of getting social care involved with meds in the NHS. If anyone knows how we can overcome the network/IG issues I’d really appreciate it :)

09:38:56
..so would the other social care systems providers (and the social care providers themselves)

09:39:37
Thanks everyone for sharing and it was very inspiring ! Great event :)

09:39:41
thankyou everyone and particularly to the all of those who have shared their expertise. Thanks to the NHSD team who have supported as well. A real demonstration of collaboration

09:39:42
Thank you all. Thank you Helen.

09:39:53
My first hackathon and fabulous to see what multiple perspectives and expertise across the sector

09:40:00
Thanks everyone!!!! Can't wait for the next one :)

09:40:11
Thank you! See you at the next one!

09:40:15
Thanks everyone