# Latest ADL workAtlanta bench and Clinical Archetype Editor **Category:** [Technical (archive)](https://discourse.openehr.org/c/technical-archive/156) **Created:** 2004-10-02 18:56 UTC **Views:** 9 **Replies:** 22 **URL:** https://discourse.openehr.org/t/latest-adl-workatlanta-bench-and-clinical-archetype-editor/12838 --- ## Post #1 by @thomas.beale Dear all, The latest versions of these two tools are available at http://www.OceanInformatics.biz. The code is open source, and a binary of the workbench is on the openEHR website, but until we make the site a bit friendlier for downloads, the packages at OceanInformatics\.biz will be easier to use\. To date we haven't had the resources to build the ADL workbench on Linux or the Mac \(which should be more or less a recompile\); these ports should be available sometime in October\. The Clinical editor will be more difficult to port as it is in VB\.net\. However I have read recently that some VB programs can be recompiled and run under mono on Linux\. We would welcome anyone with expertise on this to share some wisdom\. For Mac users, both tools will run under the Windows emulator\. This is intellectually horrible to some people, but a practical short term solution\. Both the ADL workbench and the Clinical Editor were demonstrated a number of times at the HL7 Atlanta meeting just finished \(26sep \- 1oct\), and were well received\. Clinical users are starting to see how their way of seeing things can be directly encoded into formal domain models \- the archetypes\. I presented a number of short presentations, which we will post in the next couple of days on openEHR\.org\. As usual, please post bouquets and brickbats to this list\. \- thomas beale --- ## Post #2 by @Tim_Churches > Dear all, > > The latest versions of these two tools are available at > http://www.OceanInformatics.biz. The code is open source, and a binary > of the workbench is on the openEHR website, but until we make the site a > bit friendlier for downloads, the packages at OceanInformatics\.biz will > be easier to use\. \.\.\. > Both the ADL workbench and the Clinical Editor were demonstrated a > number of times at the HL7 Atlanta meeting just finished \(26sep \- 1oct\), > and were well received\. Clinical users are starting to see how their way > of seeing things can be directly encoded into formal domain models \- the > archetypes\. Thomas, I am still a bit unclear as to whether there are plans to build an open source Archetypes/openEHR storage and retrieval server\. Personally I don't think Archetypes and openEHR will gain widespread acceptance \(as opposed to theoretical interest\) until there is some readily available means of directly using Archetype definitions to manage data\. I am aware of the Ethidium proprietary implementation, and also of the DSTC work being done for the HealthConnect pilots \- but there is no indication of an open source implementation that I can see\. There is a menu item of "EHR Server" on the openEHR web site under projects, buut the link does not go anywhere\. Note that I am not complaining that the openEHR Foundation or Ocean Informatics has not yet produced an open source Archetypes server \- it may not be part of your business plan to do so \- I am merely asking \(again \- I know I have asked this same question several times previously\) whether there are in fact concrete plans to create an open source server, and if so, what is the timetable\. We would love to use Archetypes to manage various types of public health data \- they would solve many of the data management difficulties we are running into now \- but we do not have the resources or expertise to be able to create an Archetypes storage engine ourselves\. --- ## Post #3 by @Sam Tim It is definitely in our plans to do this work\.\.\.\.it is a matter of how we get the resources together\. UCL \- with David Ingram and Dipak Kalra \- are going to take the lead on this one\. Tom will be working with them predominantly in London on this project\. Sam --- ## Post #4 by @Christopher_Clay Hi Fellows Just putting all the archive boxes in the shed and have decided to go electronic\. Have you got a system working yet or do I have to write my own? :\-\) Cheers Chris Clay \(Dermatologist\) --- ## Post #5 by @thomas.beale Tim Churches wrote: >> Dear all, >> >> The latest versions of these two tools are available at http://www.OceanInformatics.biz. The code is open source, and a binary of the workbench is on the openEHR website, but until we make the site a bit friendlier for downloads, the packages at OceanInformatics\.biz will be easier to use\. >>    > > \.\.\. > >> Both the ADL workbench and the Clinical Editor were demonstrated a number of times at the HL7 Atlanta meeting just finished \(26sep \- 1oct\), and were well received\. Clinical users are starting to see how their way of seeing things can be directly encoded into formal domain models \- the archetypes\. >>    > Thomas, > > I am still a bit unclear as to whether there are plans to build an open > source Archetypes/openEHR storage and retrieval server\. > there is\. It will be done at UCL, and prelim work is underway\. I will be spending a large amount of my time at UCL for the next 12\+ months\. We expect a first version in mid 2005 \- complete archetype and template enabled openEHR\-based data processor component; it will be deployed in initially \(probably\) a J2EE environment\. > Personally I > don't think Archetypes and openEHR will gain widespread acceptance \(as > opposed to theoretical interest\) until there is some readily available > means of directly using Archetype definitions to manage data\. I am aware > of the Ethidium proprietary implementation, and also of the DSTC work > being done for the HealthConnect pilots \- but there is no indication of > an open source implementation that I can see\. There is a menu item of > "EHR Server" on the openEHR web site under projects, buut the link does > not go anywhere\. > not today;\-\) > Note that I am not complaining that the openEHR Foundation or Ocean > Informatics has not yet produced an open source Archetypes server \- it > may not be part of your business plan to do so \- I am merely asking > \(again \- I know I have asked this same question several times > previously\) whether there are in fact concrete plans to create an open > source server, and if so, what is the timetable\. We would love to use > Archetypes to manage various types of public health data \- they would > solve many of the data management difficulties we are running into now \- > but we do not have the resources or expertise to be able to create an > Archetypes storage engine ourselves\. > well, as I say, v1 of the server by mid next year, probably earlier for initial versions of the kernel component\. It's taking a long time, but without more resources, things go at this speed\. Sad to say, vast amounts of money are continually spent on dead\-end things around our ears\! \- thomas --- ## Post #6 by @thomas.beale Bill Walton wrote: > Is the development effort going to be open to outside developers who want to > contribute? If so, how will that work? > Hi Bill, we still have to develop the roadmap for the development\. There are a number of issues, including: \- not many people \(as far as we know\) have developed an archetype\-enabled data kernel before of the kind we envisage, apart from the Ocean Informatics team; previous and current work by CHIME at UCL and by the DSTC are close\. Realistically, I don't think that too many people will want to buy into this part of the development, since the logic is quite complex\. The code will of course be freely visible, testable and if others 'get it' and want to join in, then it is just a matter of doing so\. \- there are decisions to be made about architectural deployment\. I am in favour of a route which enables a common core code base for all deployments \(J2EE, \.Net, others like Python/Zope\), but this may or may not be realistic in today's fractured world\. The process will be more or less as for any other development \- it will be visible, and interested developers can offer their input, in which case \(after a bit of conversation\) they can be added to a team\. All they have to do is agree to the commonopenEHR change management process and tools\. Where developers contribute, never having met face to face, a rock\-solid description of interfaces will of course be needed to ensure components fit together\. We will publish a roadmap for other developers to read and discuss within the next 6\-8 weeks\. I encourage people on this list who would like to be involved in development, or might have resources to offer to discuss their interests here\. I hope this answers your question\. \- thomas --- ## Post #7 by @Elkin_Peter_L_M.D Dear Thomas and Bill, Perhaps a services based architecture should be considered rather than a single "server\." By envisioning the services that we need to provide and then creating a component architecture that fits that model we may be better prepared for the future\. Also, I would recommend taking a look at the ebXML registry which is a federated Open Source registry which is currently available\. Also, Sun is implementing OWL support within the registry \(which may be handy for users interested in direct reasoning from Archetypes\)\. Warm regards, Peter Peter L\. Elkin, MD Professor of Medicine Director, Laboratory of Biomedical Informatics Department of Internal Medicine Mayo Clinic, College of Medicine Mayo Clinic, Rochester \(507\) 284\-1551 Fax: \(507\) 284\-5370 --- ## Post #8 by @thomas.beale Elkin, Peter L\., M\.D\. wrote: > Dear Thomas and Bill, > > Perhaps a services based architecture should be considered rather than a single "server\." By envisioning the services that we need to provide and then creating a component architecture that fits that model we may be better prepared for the future\. > well actually, that is our vision of course, and the services definitions we are contemplating include things like PIDS, RAD, an EHR service, and a terminology service, to start with\. I just tend to talk somewhat loosely of an EHR server from time to time, since in the background, this is what we really have to impement, along with other items of course\. > Also, I would recommend taking a look at the ebXML registry which is a federated Open Source registry which is currently available\. Also, Sun is implementing OWL support within the registry \(which may be handy for users interested in direct reasoning from Archetypes\)\. > yes, we need to get onto this together and finish the Mayo archetype/OWL work as well\! I certainly don't claim to have all the answers with regard to deployment technologies at this stage \- it will require input from all the experts here\. \- t --- ## Post #9 by @Bill_Walton Thomas Beale wrote: > Bill Walton wrote: > > >Is the development effort going to be > >open to outside developers who want > >to contribute? If so, how will that work? > > /snip/ > > We will publish a roadmap for other > developers to read and discuss within > the next 6\-8 weeks\. I encourage people > on this list who would like to be involved > in development, or might have resources > to offer to discuss their interests here\. > > I hope this answers your question\. Indeed it does\. Thanks\. I would like very much to become involved, but make no pretenses about being able to contribute as a developer\. I've been managing projects / products / programs / people for the last 15 years\. "Back in the day" I was judged to be a good technical hand \(C, Basic, and X86 Assembler\), but I've only recently begun working on getting my programming skills back \(now working on learning Java\)\. My experience tells me, though, that there's always lots of "grunt work" to be done in bringing a product to market\. Testing and documentation are two skills I can still credibly claim to have, and I'd be very happy to contribute in those areas\. Or any other areas where having some outside help would let the developers remain focused on producing working code\. Please let me know\. Best regards, Bill --- ## Post #10 by @Puvanendran_SenthilR Hi all, I am Senior software engineer specialized in Java. And worked with more healthcare solutions last 4 years and most of them in J2EE. And i have more experience on open source solutions as well. I would like to contribute with development work. Pls let me know the details. In sametime, I am doing post graduate studies as well now and would like to do my research also in this health informatics sector. So, It would be very helpful, if i get chance to work with this development. Thanks. Regards P.Senthilruban --- ## Post #11 by @thomas.beale Bill Walton wrote: > Indeed it does\. Thanks\. > > I would like very much to become involved, but make no pretenses about being > able to contribute as a developer\. I've been managing projects / products / > programs / people for the last 15 years\. "Back in the day" I was judged to > be a good technical hand \(C, Basic, and X86 Assembler\), but I've only > recently begun working on getting my programming skills back \(now working on > learning Java\)\. > don't worry I know how you feel\. My background was in C, Motorola assembler and real\-time kernels\.\.\. > My experience tells me, though, that there's always lots of "grunt work" to > be done in bringing a product to market\. Testing and documentation are two > skills I can still credibly claim to have, and I'd be very happy to > contribute in those areas\. Or any other areas where having some outside > help would let the developers remain focused on producing working code\. > You are right, and actually, having someone think about test plans and construct test cases / procedures would be a very useful thing\. We would need to dicuss how to do this exactly \(in my background we documented such things as IEEE test plan/test procedure/test cases \- for system testing that is\)\. Even just basic testing on various machine architectures we don't have access to, or debugging help pages \(or maybe even writing them?\) \- there are many many things that need doing\. We will do our best to organise things just enough to make best use of other's time and resources\. Personally, I don't underestimate the value of any of these tasks, and am sure the varied skill\-base and knowledge\-base represented by this group will help the effort immeasurably\. One of the main things is that the development is 'kept honest' for clinicians \- i\.e\. can always be shown to deliver direct and tangible benefits to clinical practice\. \- thomas --- ## Post #12 by @thomas.beale Puvanendran SenthilRuban wrote: > Hi all, >   I am Senior software engineer specialized in Java\. And worked with more healthcare solutions last 4 years and most of them in J2EE\. And i have more experience on open source solutions as well\. I would like to contribute with development work\. Pls let me know the details\. >     In sametime, I am doing post graduate studies as well now and would like to do my research also in this health informatics sector\. So, It would be very helpful, if i get chance to work with this development\. > Thanks\. > Regards > P\.Senthilruban > > \*//\* As a general piece of news: a company in Sweden, Acode, has offered a cut of java classes which implement the openEHR reference model\. The code looks very good, and we are in the midst of organising how it will be inforporated into the openEHR code base, which is maintained on public the BitKeeper server \(same server as the Linux kernel in fact;\-\) THis work could be the initial code base\. So anyone will be able to download it \(as they can already\)\. But I think a key thing is to have a roadmap of tasks and components, so that we can work efficiently and effectively\. So we want to spend a bit of time developing that\. We also have some remaining wok on the specifications \- as you can see from the CRs page \- we have to create release 1\.0 of the openEHR specifications\. I will endeavour to initiate discussions on these items as part of the review process, to give this community a chance to provide expert input, as it has in the past\. \- thomas beale --- ## Post #13 by @Bill_Walton Hi Thomas, Thomas Beale wrote: > Bill Walton wrote: > /snip/ > > > You are right, and actually, having someone think > about test plans and construct test cases / procedures > would be a very useful thing\. Excellent\. Count me in\. > We would need to dicuss how to do this exactly Development of the Test strategy should always, IME, be given explicit attention\. Would you like to have that discussion here or offline? Also, please let me know if it would be helpful to you and/or the team to have a copy of my resume\. I'll be happy to send it along\. /snip/ > or debugging help pages \(or maybe > even writing them?\) Count me in here too\. On either or both fronts\. I look forward to participating in these and any other areas where I might be able to provide some small value\. Best regards, Bill --- ## Post #14 by @Puvanendran_SenthilR Hi Thomas, Thanks. I will check them and start looking with inital codebase. Pls let me know how our team going divide the work within us. I would like to suggest some XP (eXtreme Programming Programming) tools such as Xplanner to task allocation. It really useful when team working on remote basis. [http://www.xplanner.org/](http://www.xplanner.org/) Thanks Regards P.Senthilruban --- ## Post #15 by @Bill_Walton Peter, Elkin, Peter L\., M\.D\. wrote: > > Also, I would recommend taking a > look at the ebXML registry which is > a federated Open Source registry which > is currently available\. Also, Sun is > implementing OWL support within the > registry \(which may be handy for users > interested in direct reasoning from > Archetypes\)\. Sorry for not responding to this sooner\. In all honesty, I didn't 'get' your point at first\. Heck, I may still not be getting it ;\-\) But having given it some thought and \*trying\* to understand it, I do have some thoughts and questions\. Before I get started, though, let me ask for clarification on a couple of things\. As I read the ebXML\.org site, they do not have a registry currently available\. What they have available is a \*specification\* for a registry\. The way I read the site, they expect others to actually implement registries\. If I've missed something, could you point me to it? Were you citing the ebXML registry as a mechanism we should look at using? If so, in what way? Again as I read the site, what the ebXML registry registers is companies and information about them that allows other companies to identify them for the purpose of doing business with them electronically\. Sellers publish their "Collaboration Protocol Profile or CPP as defined by \[ebCPP\] to the Registry\. The CPP describes the seller, the role it plays, the services it offers and the technical details on how those services may be accessed\. \.\.\. The buyer browses the Registry \.\.\. to discover a suitable seller\. For example the buyer may look for all parties that are in the Automotive Industry, play a seller role, support the RosettaNet PIP3A4 \[Purchase Order business protocol\] process and sell Car Stereos\. The buyer discovers the seller's CPP and decides to engage in a partnership with the seller\." \(p\. 16, "OASIS/ebXML Registry Services Specification v2\.5\) The definition and registry of the \*protocols\* \(think Archetypes\) used by the companies registered within an ebXML registry is a seperate issue, handled by organizations like RosettaNet\. It seems to me that the next issue we'll face \(perhaps very soon\) WRT Archetypes is more like the one tackled by RosettaNet than the one tackled by ebXML\. The ebXML\-type provider registry looks like an issue that won't come up until after the first one is resolved\. I'd apprecite hearing your thoughts\. Best regards, Bill --- ## Post #16 by @Real_Gaston Hi, We recently developed an EHR software in my company and although we evaluated using openEHR at the beginning of the project, we discarded the alternative due to the lack of any available real implementations of the model\. We used a similar but less complex and less powerfull approach the than openEHR model and archetypes approach, because we couldn't validate how the latter will perform in a real operational enviroment \(scalability and performance concerns\) without a reference implementation\. For this reasons, I´m really looking forward to see any advances in this subject, and to collaborate in any way I can in the development efforts\. Are this java sources available for download? Where? Thanks\. Gastón Real\. --- ## Post #17 by @thomas.beale Real Gaston wrote: > Hi, > We recently developed an EHR software in my company and although we > evaluated using openEHR at the beginning of the project, we discarded the > alternative due to the lack of any available real implementations of the > model\. We used a similar but less complex and less powerfull approach the > than openEHR model and archetypes approach, because we couldn't validate how > the latter will perform in a real operational enviroment \(scalability and > performance concerns\) without a reference implementation\. For this reasons, I´m really looking forward to see any advances in this > subject, and to collaborate in any way I can in the development efforts\. > Are this java sources available for download? > there will be soon \- in a matter of weeks\. An initial cut of Java has been done by the Swedish company Acode; the UCL team in London have been reviewing it, and I think that Acode will probably produce another cut before it goes into openEHR \(e\.g\. to include Java generics and such like\)\. So it's just a matter of the time to do this, then organise the BitKeeper source control details, and get it up on the server, at which point it will be announced\. We are also getting Bugzilla up and running for openEHR\. In the weeks leading up to the end of the year, we \(myself working with the original core team in Australia and UCL\) aim to get a development roadmap written, which can be published for discussion here\. I know lots of people would like to get right onto coding, but I cannot emphasise enough the value of planning and the large amounts of time it will save the community as a whole\. This plan has to be cognisent of the fact that various members of the community have their own agendas, e\.g\. to build a certain kind of application or tool, or to do some research, or run some kind of clinical trial, perhaps win contracts within their national health system and so on\. So it will aim to be as modular as possible\. I would suggest that programming activities might start in earnest in the new year rather than right now \- from the point of view of making the best use of others' efforts as well as one's own\. \- thomas beale --- ## Post #18 by @Philippe_AMELINE1 Hi to all, Odyssee keeps on, and our "Lignes de vie" server is now \(quite\) ready ; quite means that the code is wrtitten and the server works, but it is not tested enough to host genuine patients\. We are working hard on that\. I am convinced that openEHR and Odyssee are working on very similar concepts\. We are building the very same kind of building, except that the "concrete" used to keep things together is based on Archetypes in openEHR, while it is based on an ontology inside Odyssee\. However Odyssee already as its specific Archetypes engine, and openEHR could evolve toward an integrated ontology to acertain a common semantic\. Till now, both project are evolving side by side, with just enough knowledge of each other to "take what seems good" ; I think that we could join efforts in a more elaborated way\. The best way I can explain it is that Odyssee currently has two components running : a smart client, and a continuity of care server, while openEHR is working on health organization servers\. Put together, we get a complete system\. Regards, Philippe AMELINE --- ## Post #19 by @Sam Philippe, I agree \- your work has been an inspiration to us to keep going with the approach we have chosen\. There is no doubt that we need an ontological underpinning in our work \- and its relation to the ontological underpinning of terminology should be clear\. For a start we are populating archetypes that are discreet and can be specialised for use in different situations\. It is possible that the more terminological approach you have taken would be the best link from the archetype to pure terminologies such as SNOMED and ICD\-10\. I am interested in where you see the ideal line and where your work sits with SNOMED\-CT\. Cheers, Sam --- ## Post #20 by @Tim_Churches Philippe, Is a description of the technology and architecture used to implement the "Lignes de vie" server available? I looked on www\.nautilus\-info\.com but couldn't see anything, but then my French is very poor and I may have missed it\. --- ## Post #21 by @Philippe_AMELINE1 Hi, Since nobody on the list seems to speak french \(some kind of satanic langage however ;o\)\) \), I will try to find some time writing a text in english \(anyway the french text on the nautilus site is absolutely not up to date and I can't write in Latin \!\)\. Thanks for your interest\.\.\. Philippe --- ## Post #22 by @Walter_Dierckx Hello Philippe, bonjour, Someone here is \-indeed\- listening to your French reporting; we \(in Belgium, but by no offence, Flemish co\-operators\) are also involved in openEHR\. From within CRONOS we are \(trying\) to set up a link between the HL\-7 v\. 3 RIM and the archetypes\. Our industrial partner is Oracle, and we are trying to deploy their "Healthcare Transaction Base" \- HTB \- as a core engine in Hospital Information Systems\. Feel free to contact me regarding this issue\. Kind regards, Walter\. walter\.dierckx@cronos\.be --- ## Post #23 by @Tim_Churches Thanks\. Personally, I always thought that openEHR should have been implemented using Esperanto for all of its syntactic keywords\.\.\. --- **Canonical:** https://discourse.openehr.org/t/latest-adl-workatlanta-bench-and-clinical-archetype-editor/12838 **Original content:** https://discourse.openehr.org/t/latest-adl-workatlanta-bench-and-clinical-archetype-editor/12838