Support of SMTP on NHSnet is a great step forward as this is the standard that everyone else uses.

There is nothing wrong with X.400. There is nothing wrong with 8 track car stereos, betamax and OS2. It just that the rest of the world chose to use something else !

The rest of the world, led by the academic community, has gone for SMTP - The Internet mail standard.

In fact there are some problems with X.400 as it has been implemented in the NHS but none of these are insoluble. But It can't be described as more reliable than a typical ISPs SMTP service.

X.400 has some technological benefits over SMTP. Particularly in relation to it's built in message tracking and acknowledgement mechanisms. However, given the millions being invested in the Internet it seems unlikely that it keep this lead for long and there are ways of adding these features on top of SMTP in the interim (these have been the subject of discussion with the profession with the and the technically knowledgeable in the supplier and NHS IT community we are comfortable we have a way forward).

There is no way in which X.400 provides more or less confidentiality than SMTP. Neither have built in encryption both can easily have encryption added.

Using X.400 inevitable takes the NHS down a cal-de-sac and a rather expensive one. X.400 software is relatively expensive. Basic SMTP clients come free with windows, Unix and most other platforms more sophisticated components (say to provide a mail gateway for a practice come in ant about £40 - in one off volume!!) . There a vast choice of SMTP components for suppliers to use and lots of finished product for end users. There have been some recent attempts to offer X.400 software to the NHS at lower prices (only 10 times what similar SMTP components would cost) . I smell desperation from the X.400 software vendors to sell the stuff before it hits the car boot sales.

X.400 mail services also cost more and typically charge by the message (5p). The costs may now be hid in top slicing but there are still a cost to the NHS.

Problems which have been seen with NHS X.400 implementations (not just NHSnet but other X.400 services used by the likes of quarry house and IMC) are:

Unreliable interworking with the Internet (lost, delayed and bounced mail)

Poor handling of file attachments.  They either get destroyed or loose the file extension making using them difficult rather than just a matter of a mouse click.

There is much argument about RUA's vs MTAs ( a more sophisticated way of doing X.400). The real argument is between X.400 and SMTP.

To help you (I hope)

Those who support MTA's seem to put a number of arguments forward.

1 An MTA is the only sensible way to deliver X.400 email to the desktop.

This is true for X.400 email but we are suggesting using SMTP for email to the desktop. I think many of those in the field who support MTAs would change their views if they understood this, but nobody has told them.

2 An MTA allows you to push an urgent message to a practice.

True but so does SMTP (not POP3). In reality a practice will configure its system to dial at regular intervals during the working day to send and collect email. This is much cheaper that initiating a call from the MMHS every time a message arrives and will deliver 99% of messages more than soon enough. The rate limiting factor is the availability of the recipient and the frequency in which they look at their email not the speed of delivery. For those very few message where instant delivery is important (like certain critical abnormal lab results) I am sure that the sender will use the phone, even if the email can force delivery within a few tens of seconds, as in these situations person to person confirmation not just of a message being read but of its significance being understood is clinically essential.

This is not a MTA vs SMTP issue, both can support a push from the MMHS, but the cost and benefits of such a facility probably don't stack up.

3 An MTA give the sender certainty that their message has been delivered to the GP practice.

True. Such facilities are not integral to SMTP but are to X.400. However, it is perfectly possible to provide delivery notifications, non-deliver reports, and read acknowledgements on top of SMTP.

Acknowledgements in EDI need to be end to end and this has to be handled at an application level and this is best done in a manner independent of the email transport layer. I have had useful discussions about how we achieve this and I have committed to delivering supplier support to implement EDI acknowledgements.

I understand that the current configuration of the MMHS is to push all messages to GP MTA on receipt if this is true it is likely to generate one ISDN call per message (standard rate 4.3p). It can't be acceptable to allow this and if the MMHS can't be configured to batch messages is a overwhelming block to the use on MTA on dial-up ISDN connections.

I also feel that there may be commercial and contractual issues related to tariffs and existing contracts which my be contributing to the otherwise inexplicable support for X.400 by NHSTB. If such issues exist they must be exposed.


-------------------------------------------------------------------------------------------------
Ewan Davis - Personal email address [log in to unmask]  Also at [log in to unmask]

PGP key available from ldap://certserver.pgp.com

Work Phone numbers
Voice +44 (1) 527 579414 Fax +44(1)527 833188

Affiliations:

Chairman AAH Meditel (www.meditel.co.uk) supplier of primary care information systems.
Chairman PharMed (www.PharMed.org.uk) Promoting GP-Pharmacy links
Chairman CSSA Primary Care Group (www.cssa.co.uk) the trade association of GP System suppliers.


-----Original Message-----
From:	Paul J Scott [SMTP:[log in to unmask]]
Sent:	Tuesday, November 03, 1998 11:04 PM
To:	[log in to unmask]
Subject:	X400 and NHS Net-what are _you_ up to?

In Somerset we are starting to look at the technical aspects of
electronic comms.  What can people recommend in the way of reading
around this subject please?

And more importantly, what are different areas doing to develop their
electronic comms?  At a meeting today, the following questions were
discussed:

Why is SMTP mail being installed on NHS Net when X400 which is already
installed seems to be more reliable, more confidential and gives notice
that mail has been received?

How will individual practices route their NHS Net mail.  What are the
pros and cons of having the mail router based in BT Syntegra, in the
Health Authority, in the PCG or where else?  Should the X400 address of
a practice be based on the practice name, on the PCG name or something
else?  Is there a national policy on this?

ISDN connections will be made to all PCGs and practices joining NHS Net.
The minimum available ISDN connection is ISDN2 (2 pipes of access
between surgery and telephone exchange).  PCGs will have a permanent
connection with NHS Net down one of the pipes (68K is considered
sufficient so the other 68K pipe won't be used for the NHS Net
connection).  What happens to the other pipe?  Can it be used for the
PCG voice phone connection, or for a fax connection?  Can an outside
hacker (presumably using an ISDN router at their end) connect into the
PCG via the non NHS Net pipe, and then connect out to the NHS Net down
the NHS Net pipe?  What are the similar implications for a big practice
that already has an ISDN30 connection mixing data and voice connections.

You can see from this that the meeting produced considerable confusion,
particularly in me!  All help with clarification would be gratefully
accepted, as would your experiences in your patch.
-- 
Paul J Scott, Primary Care Physician, United Kingdom.
Fax 44 (0)1935 410188