Engineering Instructions
A History


By W.E.H. Kennedy
May 1938


I have no doubt that, to many of us, most official instructions appear to be only a necessary evil - and perhaps not even a necessary one. But from the time when Moses descended from the mountain of Sinai with the tables of the Law in his hands, every community of men which has worked together for a common end has required codes of rules, simple or complex according to its organization, to govern its proceedings. The Engineering Department of the Post Office is no exception and it is obvious that some 52,000 men of various grades and types, engaged in varied and complex functions, and scattered all over the country, require a correspondingly comprehensive code of rules.

These rules, which are contained in a number of different publications, can be divided broadly into three main classes:-

  1. Technical Instructions, i.e. descriptions of actual methods of carrying out specific working operations:
  2. Procedure Instructions, i.e. those which deal with standard forms of accounting, correspondence, records and the like:
  3. Regulations as to conduct, conditions of service, pay and privileges.


THE HISTORY of the Instructions issued by this great and growing Department is in itself a fascinating subject if we bear in mind that they are not a dead or dry-as-dust collection of immutable laws, hut a living part of the organization which faithfully reflects the changes and developments, the growths and setbacks, in the Department’s history. A general survey of this aspect of the subject would occupy too much space, but there are one or two interesting details which merit description.

The Instructions relating to accounting and other procedure and to conduct, conditions of service and the like, date, of course, from the very earliest days of the Department and were originally (as they still are to a very large extent) adaptations of corresponding rules of other Departments.

It is a little surprising to find however that there were very early attempts to provide a definite code of Technical Instructions. As early as 1879 a number of such Instructions existed and it is interesting to note that they were of exactly the same size as our present Engineering Instructions, were very similar generally as regards layout and also embodied some of the standards of spelling recently laid down for Engineering Instructions. As a little additional proof that there is nothing new under the sun, Figure 1 shows a small card instruction, extant about 1890, which can fairly be regarded as the ancestor of the new Notes for Workmen - it was evidently intended to be portable.

To dispose finally of our early history, it may be said that between 1880 and l89l there was in existence a series of sixteen Technical Instructions, prefaced by a list of definitions of technical terms. These were printed as foolscap leaflets and grew eventually into the old style Technical Instructions as we still remember them.


MEANWHILE other codes of Instructions were growing, dividing and being elaborated. The Engineering Department Regulations, containing most of the principal rules of conduct and of procedure, went through successive editions in book form until, in 1921, a beginning was made with their issue in sectionalised form. Other publications were issued as the need arose:
the principal rules of conduct and the like were printed separately in a booklet called Rules for Workmen for distribution to the workmen grades. Wayleaves, Stores Instructions and other subjects were covered in separate books or booklets in a variety of shapes, sizes and styles and the need for rapid or temporary changes in methods of procedure let loose a spate of circulars. Educational needs were catered for by an admirable set of Technical Pamphlets for Workmen. Sometime prior to 1910 the issue was commenced of the Engineer-in-chief’s Monthly List in some attempt to coordinate the various publications and give details of their issue and amendments.

By 1929 there was no lack of informative matter on any subject under the sun which affected the Engineering Department - if only one knew where to find it. The trouble was that although each of these publications dealt with a specific subject, none of them contained fixed and finite laws but had to cope with rapidly changing methods and conditions. It is safe to say that no book got to press without being pursued by amendments and quite often whole sheaves of them were published with the book. The method of amendment was simple - by gummed slips. This is an excellent method from one point of view. It is cheap, easy and effective provided that only a few amendments have to be made before a book has to be reprinted, but the disadvantages were soon only too evident. Figure 2 shows a copy of “Wayleaves” (Pg 47) open at a place which has suffered amendment and re-amendment by gummed slips. This copy is in fact the standard copy of the Engineer-in-Chief’s Editorial Section and has therefore been maintained under practically ideal conditions. The state of copies kept under less favourable circumstances can well be imagined.

The trouble did not cease with untidy and confusing books. Matter was often published in the most convenient publication from the point of view of the Headquarters author without very much regard to the convenience of the unfortunate user. For example, a considerable amount of technical matter crept into Rules for Workmen (which was intended primarily as a Staff Rule Book) because, presumably, it was at the time the handiest place to put it. Even when an apparently clear and succinct rule was available somewhere an officer, before acting on It, had to consider the possibility of amending, cancelling or contradictory rules existing in other publications, in the Monthly List, or in a mimeograph circular.

Besides this it was obvious that with Instructions issued in book form and corrected by means of gummed slips, officers interested only in one aspect of the subject had to receive and to maintain the whole of the rules relating to it.

The only way an Instruction which had become clogged with amendments could be made serviceable was by reprinting it including, of course, the accumulated amendments; the bigger the mess it was in, however, the greater the task of re-writing it became and as it was rarely that anyone could spare the necessary time, the immediate situation was dealt with by the issue of some more amendments; a thoroughly vicious circle was thus being created.


THIS THEN was the position in 1929. In that year the Telephone Section of the Engineer-in-Chief’s Office were concerned at the amount of work involved in the issue of new Technical Instructions and in re-writing those which were out of date, so far as they were concerned; and they set up a small Committee to examine the problem. As a result of this it was decided to centralize the work in a definite Group of the Section instead of leaving it scattered throughout the whole of the Section. Fortunately the Committee did not stop at this but proceeded to examine the general question of Instructions, the difficulties created by the cumbersome method of amendments, and the complications arising from amending circulars etc., as a whole problem; ultimately the matter was transferred to a Departmental Committee which reviewed the scheme completely.

It was already known that the Bell Telephone Co. of America had an excellent loose-leaf code of Instructions and the major features of this scheme were adopted. By 1930 the series of loose-leaf Technical Instructions had been commenced. The chief features of the scheme are by now familiar but perhaps the most outstanding are its flexibility in that it can cope with Instructions relating to any subject or a part thereof and the making of all amendments, additions etc. by the reissue of complete pages, so that any set or part of the set which has been properly maintained is always up to date. In addition, the sub-division of subjects makes it possible to ensure that officers receive (within reason) only those Instructions with which they are really concerned. The advantages of the system already in use for Technical Instructions were realized when, in 1934, the Regulations Committee extended the scheme to embrace non-technical Regulations. At the same time it was decided, with regard to Rules and Regulations of the Engineering Department, that all kinds of Instructions then scattered in various books, circulars etc. should be brought into the loose-leaf scheme of Engineering Instructions. This work (which is still not yet complete) involves a considerable amount of collation and simplification of matter before it can be transferred into the Engineering Instruction scheme.

It was also decided that any Instructions which for any good reason are not included in Engineering Instructions should be issued in loose-leaf form and conform as to size either with the Instructions or (for portable matter) with an octave loose-leaf pocket size.

The Engineering Instruction scheme then covered eighteen main Divisions of subjects, sub-divided into eighty seven Sub-divisions (one more Sub-division has since been added). These Sub-divisions are in turn split up into some 550 alphabetical Sections with scope, of course, for many more. Within the Sections a simple four figure numerical code applies.

It is desirable to consider how this scheme of designation was agreed to. The original Committee of 1929 which set up the loose-leaf Technical Instruction scheme considered the adoption of the Dewey decimal system of classification by numbers only, but it decided that a combination of subject classification and numbering would be more self-explanatory. The main framework (Division and Sub-division). consists of familiar titles and the Section letters and numbers, although having a significant arrangement, are secondary features, chiefly of assistance in filing. While providing (for all practical purposes) as much scope for expansion as the Dewey or any similar numerical system, the use of Division and Sub-division titles makes it possible to trace the location of an Instruction directly without translation into an arbitrary number code.

The next step was to standardize the layout and style of all Engineering Instructions and it is convenient here to consider grounds upon which the main features were adopted:-

(a) Size, demy-quarto (8.5 in x 11 in), contains approximately the same amount of press letter as foolscap but the additional width is better for tabular matter and diagrams. it is also better for two-column printing, the advantage of which in reducing eyestrain has been established by exhaustive medical research. It will be, of course, appreciated that the range of travel of the eye in reading shorter columns is less fatiguing than that resulting from full page setting.

(b) Type is now standardized at 10 point for printed Instructions in a normal reading fount; rotaprinted Instructions are produced by varitypers to conform as nearly as possible in appearance to printed Instructions.

(c) Headings and titling are standardized, and spelling is laid down to be in accordance with the Concise Oxford Dictionary, which is the recognized authority for all Government publications.

(d) Style should be as simple as possible, no foreign words being used where English ones exist and the shorter word is always preferred to the longer word.

Most of these points are dealt with in detail in "Notes for the guidance of authors and editors of Engineering Instructions" which is issued as GENERAL, General, A 0020.

The organization in the Engineer-in-Chief’s Office to effect all this is controlled by the Editorial Section which includes a technical and non-technical group: the actual work is not always susceptible of exact division on these lines and the two groups therefore work in close collaboration. Generally speaking, an Instruction whether new or amended, is received spontaneously from the Headquarters Branch concerned with the subject and the officer in that Branch who originates the draft is known as the author. On reaching the Editorial Section the draft is allotted to an officer either in the technical or the non-technical group as is most suitable and this officer is known as the editor; editors in each of the two Groups of the Editorial Section specialize in a number of subjects. The editor is responsible not merely for bringing the draft into line with standards of layout and style etc., he also has to consider what other Instructions may be affected and may consequently require cancellation or amendment, the possibility of incorporating existing Instructions in obsolescent series (such as circulars); the arrangements for distribution and their possible effect upon the scope of the Instruction and generally he has to control the production of the Instruction through its stages of printing or reproduction. These things can of course only be carried out successfully by close co-operation between the author and the editor and an appreciation on both their parts that the test of a good Instruction is its ultimate convenience to the officers who have to use it.


WITH ALL the undeniable advantages of the loose-leaf scheme however it was soon realized that workmen especially were receiving far too many Instructions with which they had no real concern. The task of maintaining files of these Instructions became therefore so onerous that it was, in many cases, quite frankly abandoned altogether. The reason was not far to seek. The original scheme provided for the Sub-division to be the basis of distribution or, in other words, for an officer needing any Instructions in a particular Sub-division to receive everything issued in that Sub-division, under the circulation letters applicable to his grade. As the scheme grew this resulted in a very large quantity of unnecessary Instructions reaching nearly all holders of files. The first step towards reducing this trouble was taken experimentally in two Engineering Sections in 1936. It consisted simply in reducing the distribution basis to the Section and it is arithmetically obvious that a considerable reduction resulted in the quantities distributed. It involved also, however, much more detailed distribution records at the local centres as it will be appreciated that the combination of 550 Sections with ten distribution letters gave a possible field of variation too wide to be recorded conveniently in schedule form. This difficulty was overcome by the introduction of a punched card system whereby selection was made semi-mechanically. This system (known as the Paramount System) is not now in use for distribution purposes and I do not therefore propose to describe it in detail here. It is sufficient to say that its application is limited by the fact that the edges of the cards only can be used for selection purposes and that therefore the larger the field of selection the larger the card must be, until a point is reached where the size of the card makes the scheme unsuitable. In addition, it was found that distribution by Section instead of by Sub-division did not wholly solve the problem and it was therefore attacked from another angle.

It was decided that the Engineering Instructions a man or a group should receive would be determined by the nature of the duties concerned and not by the subjects of the Instructions. That is to say that the user’s function instead of the subject of the Instruction was to be the deciding factor in distribution.

After an analysis of the various operations of the Department it was found that they could be divided into about 150 primary or “basic” duties. A basic duty does not necessarily represent a complete load for any man or group and such a load may comprise from one to forty or even more basic duties, according to the variety of the work performed.
The next step was to decide what basic duties were proper to every man or group receiving Instructions arid at the same time to determine, for every Instruction issued, what were the basic duties it covered. These combined tasks were of some magnitude but have now been completed.

When an Engineering Instruction is issued or reissued, the basic duties to which it is applicable are determined and copies of the Instruction are sent only to those people whose working loads include one or more of these basic duties. It was soon realized that the maintenance of the necessary records and the selection of the Instructions on this basis could only be done satisfactorily by mechanical means at a central point. The machines used are Hollerith Tabulators, adapted for the special requirements of the scheme.

The picture above shows the basic records of the scheme. The form (TE 2084) shows the basic duty numbers applicable to a particular file; it is prepared locally by a responsible officer and scrutinized at Headquarters. With it is associated the punched card on to which the particulars are transferred for machine operation. These particulars are:-

  1. Centre No., File No., and No. of copies required (on the first part of the card)
  2. Basic duty numbers (on the second part of the card)

One of these cards is prepared for every file in the country (there are between fifteen and sixteen thousand of them) and are kept amended as any changes arise.

The above picture shows the "Hollerith" machine which selects the Files to which an Instruction is to be distributed. When an Instruction is to be issued the machine is set to operate only when a card passes through it which bears one or more of the basic duties proper to the Instruction. All the cards representing all the files in the country are then passed through the machine and when the holes in the second part of the card represent any of the basic duties for which the machine is set, electrical contacts are made which cause the printing mechanism to operate and to print on a Distribution List the information punched on the first part of the card, i.e. the Centre No., File No. and No. of Copies Required.

At the same time, the divided counter in the machine records the number of copies required. When the cards for a Distribution Centre have been run through the machine the operation of a key causes the total number of copies for that Centre to be printed at the end of the distribution list and also carries this total forward in the counter, so that at the end of the run the grand total of copies for the whole country can be printed. The distribution list for each Distribution Centre is sent to that Centre direct with the proper number of copies, which are then distributed locally to the files whose numbers are shown on the list.

The picture above shows the feeding mechanism, whilst below is the printing mechanisms of the machines.

The cards are passed downwards through a narrow gate by an up and down movement of a cross beam, passing banks of wipers as they go; the position of the holes therefore determines which contacts are operated. After passing
through the machine they are stacked in a bundle just below, in their original order, and returned to their storage racks in convenient bundles. They are not therefore sorted by the machine and being always in their correct numerical sequence can easily be traced when required for amendment.

A general view of the plant is shown above. The machine on the right is engaged in making lists of Engineering Instructions appropriate to various files, which is a preparatory step in the introduction of the new Distribution Scheme. The other machine is producing an ordinary Distribution List; the large sloping rack near it contains a set of cards representing all the files of Instructions in the country. In the background are girls engaged on punching cards for revisions or replacements.


IT WILL BE APPRECIATED that the change in the basis of distribution from the subject of an Instruction to the function of its user is really revolutionary. It is still possible, of course, for an Instruction - especially a comprehensive Instruction covering a large number of basic duties - to reach a number of users who are only directly concerned with a few paragraphs. The next step, therefore, is to consider whether the actual compilation of Instructions should not take into account, more fully than at present, the functions of the user - or in other words whether Instructions should be deliberately written around basic duties. Such a change would, of course, simplify distribution but it might deprive Instructions of any real interest. There is something to be said perhaps for knowing somewhat more than only the bare details of one’s own part in a job. This, however, is the next problem for consideration.

It can readily be realized that at many points Engineering Instructions are directly connected with Regulations etc. issued by other Departments and arrangements exist for prior agreements in advance of the issue of such Instructions. This is especially true of Telephone Service Instructions and of certain circulars issued by the Telecommunications Department, but it is also necessary to secure that -any Instructions dealing with Staff, Stores or Accounting procedure are kept in step with those issued by other Departments dealing with such subjects. This need results in a considerable amount of duplication of Instructions or rather the simultaneous issue to different Departments of the same Instruction in different forms, and I shall recur to this point later.


ALTHOUGH it is desirable to include, in the Engineering Instruction scheme, all Instructions which can usefully be fitted into it, a certain number of special Instructions remain outside. These are kept separate for one or two reasons - either because they must be carried on the person and therefore reproduced in a smaller size (such as the Precautions against Accidents and the Works & Maintenance Units) or because they cover special subjects which require specialized distribution; in the latter category come the Rules for Workmen (which are supplied personally to every workman); the Regulations for the Cable Ships (which concern only the few people employed in the ships); the Rules for Motor Transport Workshop staff, and the Instructions to Underground Works Supervisors.

It is possible, however, that the new Distribution Scheme may enable some of these to be incorporated in Engineering Instructions - at any rate where special distribution arrangements are at present the only reason for their exclusion from the scheme.

In addition to the codes of definite Instructions which have a mandatory significance, there are the Educational series represented at present by the Technical Pamphlets for Workmen (now to be superseded by loose-leaf Educational Pamphlets) and the Notes for Workmen which are intended to be pocket reminders of Engineering Instructions. Clerical Training Pamphlets have been issued in connexion with the training of recruit clerical staff and are designed to explain principles rather than details of working procedure. The loose-leaf principle is common to all these publications and any which may still exist in book form are being reissued in loose-leaf form as quickly as possible.


I THINK it will be conceded that the Engineering Department has, at any rate, tried to tackle the thorny problem of its Instructions in a workman like manner and it is not surprising therefore that we have had imitators. Last year a Committee was appointed by the Deputy Director General to consider the possibility of extending the loose-leaf scheme of Engineering Instructions to other Headquarters Rule Books. As a result, a loose-leaf scheme has been adopted for Telephone Instructions, Postmaster’s Manual and the Surveyors Code. This does not follow our scheme in every particular and, of course, the distribution problems are simpler as there is not the diversity of conditions and mixture of duties which we have to encounter. Still, the important elements of the scheme have been adopted.

Following upon this - another Committee has recently been appointed to consider the concentration of all Post Office Instructions under a central editorial body. As I pointed out earlier, on some subjects the same Instruction is issued in slightly differing form by a number of different Departments and a certain amount of work is involved in getting these in agreement. Difficulties in this connexion will, of course, continue to be felt more and more by the Regions and Areas who with their combined functions have to attempt to combine the implications of a number of different codes of Instructions emanating from different sources.

The logical development of centralized publication would therefore be common code of Post Office Instructions with a system of sub-divisions and machinery for distribution designed to embrace the whole working staff of the Post Office, supplemented by perhaps a few separate codes for special purposes.

BACK Home page BT/GPO Telephones Search the Site Glossary of Telecom Terminology Quick Find All Telephone Systems

Last revised: December 11, 2021