POL00142397
POL00142397
From: Anthony de Garr Robinsori”
Sent: Wed 12/09/2018 1:13:48 PM (UTC)
To: Andrew Parsoné_
Ce: Jonny Gribben; ; Lucy
Bremnerf
Subject: RE: Structure for expert report [WBDUK-AC.FID26896945]
Dear Andy,
This draft essay plan from Robert is both timely and helpful. I set out below my and Simon’s combined
thoughts on the plan. We recognise that some of these thoughts may miss their mark and that Robert will be
alive to many of them points already, but would be interested to know what he thinks about them, and about
how and where in his report he will deal with them.
Overview
1. We like the first part of Robert’s overview section and we particularly like his positioning of
robustness as the critical issue. However, we all need to be clear that for the client, the key objective is
to satisfy the court that, when faced with any discrepancy in any branch’s accounts, the chances are very
small indeed that this deficiency is caused by Horizon, still less by any of the specific matters identified
by Coyne. That the Horizon system is well designed / built is highly relevant to and helpful on that key
question, but it is not the key question. We want to win even if (as seems to be suggested in some of the
reports and other documents referred in the outline claim document) parts of Horizon were not perfectly
designed / built for some or all of the time.
2. When we say “caused by Horizon”, we mean caused by a bug or caused by any of the other issues
raised by the common issues and/or by the outline claim document and/or by Coyne’s report. At the risk
of stating the obvious, Robert needs to ensure that he does not focus exclusively on bugs. We note that
the outline claim document also makes complaints about “error repellency” and about the ability of
Horizon to alert SPMRs to bugs (Issue 2); the adequacy of controls (Issue 4 and Issue 6); and the
general process of reconciliation (Issue 5). Does Robert need an additional section or sections
somewhere addressing the other issues?
3. In this context, it is worth adding that Robert should not simply focus on the specific criticisms which
Coyne makes. We know that there are likely to be strong responses to them, but we also know that
Coyne’s approach is to throw as much mud in the hope that some sticks so he is likely to shift ground
onto the more amorphous criticisms if he feels he needs to. The point is that the consideration of
robustness and countermeasures needs to cover all this as well.
4. While on the subject of robustness and countermeasures (and thus the chances of Horizon causing
discrepancies in branch accounts), to what extent will Robert be addressing the business processes
surrounding the operation of the Horizon system, in addition to the countermeasures that are built into
Horizon itself? If those processes are operated properly, branches should (for example) do cash balances
on a daily basis. If these are done, it seems to us that the chances of an error going undetected and
becoming impossible to resolve, reduce to a vanishingly small probability. The Horizon issues should
not be approached as if the system is entirely automatic i.e. not affected by human interaction. An
essential part of the system is how it is used by its users. To take an extreme example, if a user
(negligently or fraudulently) persistently enters £20 as having been deposited when £200 was in fact
deposited, and then fails to carry out any checks or worse, puts misleading entries in, there is little
Horizon can do about it. All this is obvious but we wonder if there is scope for Robert to address the
proper use and human operation of Horizon i.e. that part of the robustness of the system comes with
users following proper procedures on a regular basis. This certainly would not be a principal line of
attack, not least because we suspect that a significant number of postmasters may have recorded false
balances in Horizon in order to conceal shortfalls. But we would like the judge to get a feel for the point
POL00142397
POL00142397
and would be interested to know Robert’s views on whether and how it can properly and usefully be
included in his analysis of robustness etc.
5. As regards the second part of his overview section:
Yes, robustness and countermeasures should be placed up front, but as indicated in para 1 above, this
is a stepping stone to the key question, which should be clearly identified.
As to the proper scope of the robustness question and what should be treated as countermeasures, see
para 4 above.
Are Robert’s views on the high level of robustness consistent with all the KELs referred to in paras
1.4, 1.4 and 4.1-4.3 of the outline claim document and the contemporaneous reports referred to in
para 3.2(f) of that document? He needs to deal thoroughly with points such as this — as we are sure he
is aware.
Second and third bullet points — in the joint statement, Robert and Coyne both classed issues 1, 3, 4
and 6 together, but Worden now puts issue I after issues 3, 4 and 6. We are not sure that matters
very much and, in any event, we think his way of looking at issues 3, 4 and 6 means that he sees them
as stepping stones and subsidiary to issue 1, which raises the key question identified in para 1 above.
If that is right, we like it. However, he will need to explain why he has taken this slightly different
structure.
We do not like the way Robert approaches issue 1 — starting with the relevant KELs then going
quantitative (see also para 8 of the summary section at the end of his essay plan). Where does he
include his analysis on the key point that undetected bugs were very unlikely to be the cause of.
permanent shortfalls in a branch’s accounts (the point he repeatedly makes in the joint statement, in
relation both to issue 1 and to issues 4 and 6)? This is much more important than and logically prior
to the quantitative approach. It should accommodate his analysis of the relevant KELs (or at least the
KELs referred to in para 1.3 of the outline claim document), including his analysis of how long it took
to identify them etc. Hopefully, those KELs could be used as strong support for his conclusion,
depending on things such as (1) how long it took to detect those KELs, (2) which features of those
KELs resulted in their arising and not being detected until they were in fact detected, and (3) whether
there are reasons for thinking it very unlikely that other bugs with similar features would have arisen.
See further para 7 below.
As to the quantitative upper limits approaches Robert refers to:
i. We welcome the fact that he recognises the desirability of treating them as backups (in case he is
wrong on the key question).
ii, I We also welcome that he recognises the need to explain the approaches and their logic in simple
language in the body of the report, leaving any fancy calculations to an appendix.
iii. We are unclear about which approaches he will be adopting and the order in which he will be
presenting them. Can he clarify on a single piece of paper?
iv. Will he be doing different calculations for different periods, before and after significant changes
in Horizon — e.g. pre and post 2010? Does the logic of his approach suggest that he ought to
do a separate calculation for each update/iteration of Horizon? If so, is this a problem for his
approach?
v. Four big questions occur to us, some or all of which may be stupid:
(1) First, is it possible for him to do a calculation of the chances of a bug affecting a branch’s
accounts, before he calculates the maximum likely impact of such a bug on a branch’s
accounts? If this is possible, we see this as extremely desirable.
(2) Second, is there any possible tension between his conclusion that the chances of that
happening are very low and his calculation of the maximum likely impact when it does
happen?
(3) Third, I have looked at Gareth Jenkins’ note on the payments mismatch bug dated
29/09/10 and I see that para 6 says that it had so far been identified that the bug had caused
one discrepancy of £30k, one of £4,800 and others of less than £350. Does this pattern of
losses call any or all of his quantitative methodologies into question? If not, why not?
POL00142397
POL00142397
(4) Fourth, as we discussed at our con in August, we wonder whether there is a point worth
making in this context about the (im)probability of the patterns of losses alleged by the
Claimants — which we think may be no or small losses for month after month, then much
bigger losses in a few months (we may be wrong, of course). Might this be undermined or
called into question by the Jenkins note and/or by the fact that Claimants may have
accumulated small losses over a long period and concealed them by false accounting until it
was no longer possible to do so? We would be interested to know what Robert thinks
about this.
g. Fifth bullet point — do we have the chapters he refers to, or the most up to date version of those
chapters?
Robustness: Countermeasures
6. The first part of this section makes points that seem sensible. However, we have two queries:
a. Is it a problem that some contemporaneous reports appear to suggest that there were some serious
deficiencies in Horizon (see para 3.2(f) of the outline claim document)?
b. Robert will ideally be addressing the position when Horizon was first rolled out — will he be able to
do so? More generally, will he be able to deal adequately with the position pre-Horizon online in
2010?
7. As regards the second part of this section (the table):
a. Do these points allow an assessment to be made of the (very low) likelihood of undetected bugs being
the cause of permanent shortfalls in a branch’s accounts? If so, how? Which of the countermeasures
had that effect? Did any other features that are not countermeasures also contribute to that effect
(e.g. reconciliation, facilities for subpostmasters, facilities to Post Office etc)? If so, where are they
taken into account?
b. Are all of the points made in the table (e.g. good support processes):
i. safe to say?
ii. true for the whole of the period 2000 to 2018?
iii. consistent with the existence/duration/history of (1) the three bugs referred to in para 1.3 of the
outline claim document (we would love to know how quickly they were identified) and (2) the
other bugs referred to elsewhere in the outline claim document?
iv. consistent with the para 3.2(f) reports referred to above?
c. Re- item 12 in the table, will this item include the business processes discussed in para 4 above? If
not, where will those processes be covered?
On any view, some careful reasoning/deft footwork will be needed to deal with these points.
Summary of sections of the report
8. We are not sure how and where issue 2 (alerting subpostmasters to bugs etc), or issues 7, 10, 11, 12
and 13 (remote access), or issue 15 will be addressed?
9. Para 4 —is there sufficient available material to allow Horizon pre 2010 to be addressed adequately?
And what about the possibility that Horizon was terrible when first rolled out? Also, should all Robert’s
calculations (of the low probability of Horizon causing losses, of the likely maximum amount of such
losses etc) be different for the period up to 2010 and the period after it? See para 3.e.iv above.
10. Para 8 — see our points above. The most important point is that we would like a
calculation/assessment of the probability of an undetected bug (and other Horizon issues such as error
repellency?) causing losses in branch accounts, before any calculation/assessment of the maximum likely
losses caused thereby.
11. Para 9 — does issue 10 go with issue 5?
Happy to discuss whenever convenient.
Best wishes,
Tony
Anthony de Garr Robinson QC
www.oeclaw.co.uk
ONE
One Essex Court, London
Temple
London EC4Y 9AR
United Kingdom
The contents of this email are CONFIDENTIAL and may be PRIVILEGED. If you are not the intended recipient, please telephone (020)
7583 2000 and delete this email.
From: Andrew Parsons {7
Sent: 07 September 2018 07:36
ESSE
Switchboard: i
Fax
X COURT
One Essex Court, Singapore
Maxwell Chambers #02-15
32 Maxwell Road
Singapore 069115
Tel {
singapore¢
Lil
300
LEADING SET
Chambers UK Commercial Litigation
Set of the Year 2015
POL00142397
POL00142397
POL00142397
POL00142397
To: Anthony de Garr Robinson imon Henderson ¢_
>; Emma Campbell-L
Lucy Bremner ¢ ; Gavin Matthews <. ne
Subject: FW: Structure for expert report [WBDUK-AC.I F1D26896945]
Tony, Simon
I asked Robert to put together an outline of his report so that we can start to see the overall picture. I think this has
helped move his thinking away from his quantitative analysis, back something closer to what we were all expecting.
All comments on the outline much appreciated. For my part, I thought it looked good.
A
Andrew Parsons
Partner
Womble Bond Dickinson (UK) LLP.
womblebonddickinson.com
WOMBLE
BOND
DICKINSON Yin)
From: robert worden
Sent: 06 September 2018 16:58
To: Andrew Parsons
Cc: 'Chris Emery’; Jonathan Gribben
Subject: Structure for expert report
Hi Andy -
Attached is our proposed structure for the expert report , for your comments.
Robert
Please consider the environment! Do you need to print this email?
The information in this e-mail and any attachments is confi
mail and
only is authorised to access this e+
‘achments. If you are not copies. Unauthorised use,
istribution, publication oF Copy is
Privacy Policy on our website
dissemination,
Any files attached to this e-mail will have b
Joss or damage which may be caused by sofiw
checked by us with virus detection software before transmission, Womble Bond Dickinson (UK) LLP accepts no liability for any
e viruses and you should carry out your own virus checks before opening any attachment.
Content of this email which does not relate to the official business of Womble Bond Dickinson (UK) LLP, is neither given nor endorsed by it.
This email is sent by Womble Bond Dickinson (UK) LLP which is a limited liability partnership registered in England and Wales under number 0C317661. Our registered office
POL00142397
POL00142397
is 4 More London Riverside, London, SE1 2AU, where a list of members' names is open to inspection, We use the term partner to refer to a member of the LLP, or an employee
or consultant who is of equivalent standing. Our VAT registration number is GB123393627.
Womble Bond Dickinson (UK) LLP is a member of Womble Bond Dickinson (Intemational) Limited, which consists of independent and autonomous law firms providing
services in the US, the UK, and elsewhere around the world. Each Womble Bond Dickinson entity is a separate legal entity and is not responsible for the acts or omissions of, nor
can bind or obligate, another Womble Bond Dickinson entity. Womble Bond Dickinson (International) Limited does not practice law, Please see
‘www.womblebonddickinson.com/legal notices for further details
Womble Bond Dickinson (UK) LLP is authorised and regula
(ed by the Solicitors Regulation Authority
This email has been scanned by the Symantec Email Security.cloud service.
For more information please visit http://www.symanteccloud.com
This email has been scanned by the Symantec Email Security.cloud service.
For more information please visit http://www.symanteccloud.com