Comments for Marc de Graauw - Random Notes http://www.marcdegraauw.com Some thoughts on meaning, the web and everything Sun, 27 Jul 2008 08:40:44 +0000 http://wordpress.org/?v=2.2.2 Comment on The trouble with PSI’s by David Byrden http://www.marcdegraauw.com/2007/01/28/the-trouble-with-psis/#comment-9182 David Byrden Fri, 25 Jul 2008 13:48:30 +0000 http://www.marcdegraauw.com/2007/01/28/the-trouble-with-psis/#comment-9182 >> "PSI’s are URI’s which uniquely identify something" To be more correct, PSIs are the resources that you get when you resolve the URIs. >> “PSI’s are URI’s which uniquely identify something”

To be more correct, PSIs are the resources that you get when you resolve the URIs.

]]> Comment on Implementing Healthcare Messaging with XML by Marc de Graauw http://www.marcdegraauw.com/2007/12/10/implementing-healthcare-messaging-with-xml/#comment-7863 Marc de Graauw Fri, 23 May 2008 08:28:32 +0000 http://www.marcdegraauw.com/2007/12/10/implementing-healthcare-messaging-with-xml/#comment-7863 Ravi, currently we handle all errors on their own level, so one can get http errors, soap faults, HL7 errors. HL7 errors are normal soap responses. An alternative approach is to package accept acks containing just an error code in a soap fault - I think that approach is equally valid, I can see no real problem. True, the error appears on more levels - http 500, soap fault, hl7 error, but soap already does that itself by packaging soap faults in http 500. Hope this helps, Marc Ravi,

currently we handle all errors on their own level, so one can get http errors, soap faults, HL7 errors. HL7 errors are normal soap responses.

An alternative approach is to package accept acks containing just an error code in a soap fault - I think that approach is equally valid, I can see no real problem.

True, the error appears on more levels - http 500, soap fault, hl7 error, but soap already does that itself by packaging soap faults in http 500.

Hope this helps,

Marc

]]> Comment on Implementing Healthcare Messaging with XML by ravi http://www.marcdegraauw.com/2007/12/10/implementing-healthcare-messaging-with-xml/#comment-7850 ravi Thu, 22 May 2008 20:25:48 +0000 http://www.marcdegraauw.com/2007/12/10/implementing-healthcare-messaging-with-xml/#comment-7850 Hi Marc, I went through some of your blogs and material regardfind hl7v3 implementation. we are trying ti implement PIX/PDQ for subject discovery and also document query and retrieval. I was wondering how did you take care of error codes specified in those transactions. I found a document where they mention about how error code should be handled for XDS document query and retrieval but i did not find anything for subject discovery. just another dumb questions. i am assuming that whatever error codes are specified they have to be send as either a soap response or ebrim error message but NOT as soap fault. I did not know if my assumption is true or not. it would be helpful if you can comment on it and share your experience on this. Thanks ravi Hi Marc,
I went through some of your blogs and material regardfind hl7v3 implementation.
we are trying ti implement PIX/PDQ for subject discovery and also document query and retrieval.
I was wondering how did you take care of error codes specified in those transactions.
I found a document where they mention about how error code should be handled for XDS document query and retrieval but i did not find anything for subject discovery.

just another dumb questions. i am assuming that whatever error codes are specified they have to be send as either a soap response or ebrim error message but NOT as soap fault. I did not know if my assumption is true or not.

it would be helpful if you can comment on it and share your experience on this.

Thanks
ravi

]]> Comment on The trouble with PSI’s by chris sizemore http://www.marcdegraauw.com/2007/01/28/the-trouble-with-psis/#comment-6305 chris sizemore Thu, 28 Feb 2008 22:51:08 +0000 http://www.marcdegraauw.com/2007/01/28/the-trouble-with-psis/#comment-6305 make it easy on yourself and just use Wikipedia URIs (or, if you must, dbPedia.org URIs)... you are both worrying too much... ;-) make it easy on yourself and just use Wikipedia URIs (or, if you must, dbPedia.org URIs)…

you are both worrying too much… ;-)

]]> Comment on Implementing Healthcare Messaging with XML by Marc de Graauw http://www.marcdegraauw.com/2007/12/10/implementing-healthcare-messaging-with-xml/#comment-6233 Marc de Graauw Tue, 26 Feb 2008 09:34:13 +0000 http://www.marcdegraauw.com/2007/12/10/implementing-healthcare-messaging-with-xml/#comment-6233 Marouan, I think WSRM is fine. HL7 is abandoning accept acks for reliability itself, and anyway one has to implement reliability with accept acks oneself. I have much more doubts on WSA. WSA is one the most "unRESTful" parts of WS-*, and I think we should take the REST criticism seriously (whereas REST has no generally accepted reliability solution, and WSRM is not that unRESTful). Not do I really believe the underlying view of WS-* and WSA where lots of intermediaries without knowledge of the message do intermediairy stuff, and that is the problem area which WSA addresses. So I'd preferably use Basic Profile, WSRM, and only those parts of WSA which are needed for the particular WSRM solution, not use HL7 accept acks, and keep HL7 Sender and Receiver. Hope this helps, Marc Marouan,

I think WSRM is fine. HL7 is abandoning accept acks for reliability itself, and anyway one has to implement reliability with accept acks oneself. I have much more doubts on WSA. WSA is one the most “unRESTful” parts of WS-*, and I think we should take the REST criticism seriously (whereas REST has no generally accepted reliability solution, and WSRM is not that unRESTful). Not do I really believe the underlying view of WS-* and WSA where lots of intermediaries without knowledge of the message do intermediairy stuff, and that is the problem area which WSA addresses. So I’d preferably use Basic Profile, WSRM, and only those parts of WSA which are needed for the particular WSRM solution, not use HL7 accept acks, and keep HL7 Sender and Receiver.

Hope this helps,

Marc

]]> Comment on Implementing Healthcare Messaging with XML by Marouan OURI http://www.marcdegraauw.com/2007/12/10/implementing-healthcare-messaging-with-xml/#comment-6228 Marouan OURI Tue, 26 Feb 2008 00:05:52 +0000 http://www.marcdegraauw.com/2007/12/10/implementing-healthcare-messaging-with-xml/#comment-6228 Hi Marc, It's one the important question that i am thinking on. In your perspective and after your recent chat on the conference "Implementing Healthcare Messaging with XML", do you think that the WS* (WS RM and WS addresing) are enough mature and performant to use ? Especially for a large implementation of an EHR ? Or simply use the basic WS profile with the native hl7 message capabilities like the accept ack. Thanks. Marouan Ouri Hi Marc,

It’s one the important question that i am thinking on.

In your perspective and after your recent chat on the conference “Implementing Healthcare Messaging with XML”, do you think that the WS* (WS RM and WS addresing) are enough mature and performant to use ? Especially for a large implementation of an EHR ?

Or simply use the basic WS profile with the native hl7 message capabilities like the accept ack.

Thanks.

Marouan Ouri

]]> Comment on Eat Your Own Dogfood by Marc de Graauw http://www.marcdegraauw.com/2007/12/21/eat-your-own-dogfood/#comment-6039 Marc de Graauw Mon, 18 Feb 2008 08:36:15 +0000 http://www.marcdegraauw.com/2007/12/21/eat-your-own-dogfood/#comment-6039 Dirk: You are right in saying it's not easy. I would prefer to write natively in html, but the html editors around are nowhere as easy to use as an office suite. I usually write in Word, when the doc is stable enough, bite the bullet and convert to html, then maintain in html. I've seen people use wikis for standard writing as well. The editors are not very good, but wikis have lots of advantages - like the community feedback you mention. Dirk:

You are right in saying it’s not easy. I would prefer to write natively in html, but the html editors around are nowhere as easy to use as an office suite. I usually write in Word, when the doc is stable enough, bite the bullet and convert to html, then maintain in html.

I’ve seen people use wikis for standard writing as well. The editors are not very good, but wikis have lots of advantages - like the community feedback you mention.

]]> Comment on Eat Your Own Dogfood by Dirk Temme http://www.marcdegraauw.com/2007/12/21/eat-your-own-dogfood/#comment-6038 Dirk Temme Mon, 18 Feb 2008 08:25:42 +0000 http://www.marcdegraauw.com/2007/12/21/eat-your-own-dogfood/#comment-6038 I agree. But I am struggling with that myself, when writing standards documents or manuals or so. I write the doc's in OpenOffice but then I would like to convert it easily to .pdf AND to .html. Pdf is a good option in Oo, and you can also export to .xhtml. But then there is a lot of extra work to be done. The .xhtml needs serious reformatting. And then probably one would like to give readers an opportunity for comments, or some other fancy stuff. So you need to go into xslt, php, ... A bit to much for most document writers. So I am looking for a bit better tooling here. You know some? I agree. But I am struggling with that myself, when writing standards documents or manuals or so. I write the doc’s in OpenOffice but then I would like to convert it easily to .pdf AND to .html. Pdf is a good option in Oo, and you can also export to .xhtml. But then there is a lot of extra work to be done. The .xhtml needs serious reformatting. And then probably one would like to give readers an opportunity for comments, or some other fancy stuff. So you need to go into xslt, php, … A bit to much for most document writers. So I am looking for a bit better tooling here. You know some?

]]> Comment on Axioms of Versioning 2 by Bert Oldenburger http://www.marcdegraauw.com/2007/09/12/axioms-of-versioning-2/#comment-4723 Bert Oldenburger Mon, 03 Dec 2007 15:54:59 +0000 http://www.marcdegraauw.com/2007/09/12/axioms-of-versioning-2/#comment-4723 It seems to be more a question of the right font. For example the symbol ∈ 'Element of'. I have found http://www.fileformat.info/info/unicode/char/2208/fontsupport.htm The end of this tunnel is in sight but I don't presently have enough time in my fueltank to see it through. A solution may follow &soonish;) It seems to be more a question of the right font.

For example the symbol ∈ ‘Element of’.

I have found
http://www.fileformat.info/info/unicode/char/2208/fontsupport.htm

The end of this tunnel is in sight but I don’t presently have enough time in my fueltank to see it through. A solution may follow &soonish;)

]]> Comment on Axioms of Versioning 2 by Marc de Graauw http://www.marcdegraauw.com/2007/09/12/axioms-of-versioning-2/#comment-4455 Marc de Graauw Fri, 16 Nov 2007 08:52:54 +0000 http://www.marcdegraauw.com/2007/09/12/axioms-of-versioning-2/#comment-4455 Bert Oldenburger: "This would be very interesting, if my browser would show me the mathematical symbols in your discourse not as [] boxes..." This is indeed - as you notice - a illustrating shortcoming in especially a note on a subject such as versioning. I used all HTML escapes such as ∃ for "there exists" (the reverse capital E) and ∀ for "for all" (upside down capital A), assuming this would ensure support among all browsers. Which version of which browser do you use? Bert Oldenburger:

“This would be very interesting, if my browser would show me the mathematical symbols in your discourse not as [] boxes…”

This is indeed - as you notice - a illustrating shortcoming in especially a note on a subject such as versioning. I used all HTML escapes such as ∃ for “there exists” (the reverse capital E) and ∀ for “for all” (upside down capital A), assuming this would ensure support among all browsers. Which version of which browser do you use?

]]>