The Objective II Committee gratefully acknowledges the general good guidance, encouragement and technical assistance provided by Mrs. Darleen Bogart, chairman of the Unified English Braille Code Research Project Committee (our parent committee), by the other members of that committee, by other participants in this project, by the officers of the International Council on English Braille and persons associated with its constituent national braille authorities, and by many braille readers and people everywhere who care about braille. This widespread good will and interest has positively inspired us with a sense of hope that important changes are possible, even as we all share a commitment to the stability of English Braille.
This report is the summary to date of the work of the Objective II Committee (commonly called "Committee 2"), which was originally charged with defining the basic methodology for extending the base literary code (English Braille), as the first technical step towards a Unified English Braille Code (UEBC or, more simply, UEB), a research project of the International Council on English Braille (ICEB). In 1993, Committee 2's charge was expanded to include the carrying out of the UEB code extensions to mathematics, computer science, and other technical fields.
This report supersedes any of the previous reports of this committee, including the "final" reports of November 1992 and March 1995 (Ref. 95a), the interim report of March 1994, and the supplementary report of October 1999 (Ref. 99a). It also includes the symbols introduced in the joint report (with the Objective IV Committee) of February 2001 (Ref. 2001a), although that report also remains substantially current as only one of its recommendations has been subsequently reconsidered by Committee 2 (see section 3.23 herein).
As a committee report, this should be regarded as a recommendation only, still subject to approval and acceptance by the UEBC Project Committee, by the ICEB, and by the ICEB's constituent national authorities. However, to keep the wording simple, UEB is described herein as if it were already an adopted code.
It is also described as if it were "finished," but of course that has never been true of any braille code, and UEB is no exception.
This document does not include extensive examples, using them
only to provide
immediate clarifications in some cases. Several
UEB example sets
do exist and are available on the ICEB web
site
(http://www.iceb.org).
This report departs from its predecessors in that it attempts to keep the language of the body of the report as nontechnical as possible, speaking directly to those who would potentially read and write UEB. Technical jargon and formalities (none all that deep, actually) have been relegated to the appendices.
Wherever we refer to "current English Braille" or simply "EB" herein, we mean the official literary code currently used in English-speaking countries, without regard to the minor differences that exist in the practical implementation of that code in different places (Refs. 91a, 92a).
Respectfully submitted,
The Objective II Committee:
The following people have also served as members, either regular or pro-tem, and have therefore participated in the work reported upon here:
At the time of this report, as at the time the UEB project was launched in 1992, the braille codes defined by the Braille Authority of North America (BANA) are divided according to subject area. Other than music, which follows an internationally accepted standard, those areas can be broadly described as (a) general literature not involving specialist notation, (b) math and those sciences employing similar notation, including chemistry, and (c) computer science.
For its first year, when it was strictly a BANA project, the objective of UEB was to bring those three distinct American braille codes into a unified system so that, for just one widely used example, one would not have to learn three different braille representations for the dollar sign.
As work progressed, interest grew in other English-speaking countries, including those where the codes defined by the Braille Authority of the United Kingdom (BAUK), or slight variations thereof, are used. BAUK also defined three distinct codes split along the same functional lines as BANA, although the BAUK math code, especially at the lower levels, can be regarded as an extension of the general literary code. But in any case, while the BANA and BAUK literary codes are close enough to be regarded as variants of the same code, the same is not true of the technical codes: both the math codes and the computer codes are very different in the two jurisdictions. Consequently, when the UEB project was internationalized in 1993, the concept of unification came to be applied to geographical as well as subject-matter divisions.
Within the UEB project, Committee 2 was given the job of defining just how the unification would take place, starting with an assumed "base" code that would be as much like the current general literary code, namely English Braille (EB), as possible.
For more details regarding the specific charges and guidelines given to Committee 2, and our interpretations thereof, see Appendix A.
In order to carry out its charge, Committee 2 sought to define very precisely how symbols are represented in literary braille, remaining faithful to current literary practice as much as possible while providing a basis for extension to technical notation — because in technical material especially, precise knowledge of the represented symbols and their relationships is essential to proper understanding.
That is, we considered that the basic units of written communication are symbols, such as letters, digits, punctuation marks and so on. In order to read, you, the reader, must not only recognize these symbols one by one but must also assemble them into larger units such as words and mathematical expressions, and then into even larger units such as sentences and paragraphs — in short, you must make sense of them.
But while the extraction of sense is your human task — and privilege and reward — you have a right to expect that the elementary symbols that you use as a starting-point will be clear and complete. In other words, it is the task of any symbol representation system, whether it be print or braille or anything else, to be capable of expressing any and all necessary symbols with complete accuracy — that is, unambiguously.
UEB provides such a system. Notice that, in UEB, we do not regard braille as representing print or as secondary or "slavish" to print in any way. Rather, we regard braille and print as parallel systems for representing symbols, and have designed UEB to do that job of representation just as well as print — that is, with the same expressiveness and accuracy.
If you would like to learn more about this philosophy and the other design principles that underlie UEB, together with related technical definitions and other details, consult Appendix B. In the main body of this report, we will present UEB in as nontechnical a way as possible, with the intention that you will learn how to read UEB — in other words, that you will learn "the reader rules." (And yes, the double meaning in that phrase is intentional, because readers have governed the process of UEB development throughout.) If you already know English Braille, you will find much of this to be obvious and familiar, since, as already stated, UEB was designed to depart as little as possible from current EB. (In fact, if you are reading this in English and in braille, and the parentheses around this sentence are among just a few unfamiliar symbols that you have encountered, you are already reading UEB — it is that easy.)
There are only 64 distinct patterns, including the space, that can be expressed with a single 6-dot braille "cell." Since the number of symbols that must be represented is many times that number, it is necessary for UEB to employ multi-cell symbols. A natural question arises: If you have a series of symbols, all run together, and some of them are or could be multi-cell symbols, how do you tell where one symbol ends and another begins? For UEB, the answer lies in a principle of symbol construction known as "prefix-root."
To explain the prefix-root concept, we first divide the 63 distinct braille cells (excluding the space) into two groups, prefixes and roots. Prefixes are the patterns having only right-hand dots plus the traditional numeric prefix, dots 3456. The roots are all the other dot patterns. Assuming that you know where a symbol begins, the rules for telling where it ends can then be given as follows:
1. If the first symbol is a root, or the space, that's also the end — in other words, it's a one-cell symbol. For example, all the following are complete symbols, and have the basic (grade 1) meanings indicated:
2. If the first symbol is a prefix, then continue until a root or a space is encountered. In the former case, the root completes the multi-cell symbol and is included in it. The vast majority of UEB symbols are of this type. In the latter case, the symbol being read stops short of the space, and therefore comprises only prefixes. Such prefix-only symbols may only occur before spaces and so have very limited application, namely to certain braille indicators that make sense only at spaces. (An indicator is a symbol that does not in itself directly represent a concrete or "printable" character, but that tells you something about one or more nearby symbols — what they mean or how they are formatted, for example.) Examples of symbols that could be read following this rule are the following:
3. To avoid major departure from prevailing custom, there are a very small number of exceptional symbols that do not fit within these simple rules (or, if you prefer, they fit within a somewhat more complex set of rules as given in Appendix B). There aren't many of these exceptional symbols that have been assigned, and it's unlikely that many more will be, and so the easiest way to deal with them is simply to list them:
These are recognized on a longest-first basis. For example, dots 56, 56 followed by something else would always be considered a grade 1 word indicator, not two instances of a grade 1 symbol indicator.
Note that all of these exceptional symbols are indicators, and comprise only prefix cells. However, unlike other prefix-only symbols, these symbols may be and typically are used before symbols other than space. Even so, by virtue of transcribing and symbol-assignment rules that are spelled out in detail in Appendix B, the existence of these prefix-only symbols do not give rise to symbol-boundary ambiguities no matter what symbols may follow them. A proof of this fact is given in Appendix C.
So, with these simple rules in mind, you can always tell where one symbol ends and another begins in UEB, even in those cases where you encounter a symbol that is new to you and need to look it up. At least you know what to look up!
All that remains, then, is to list the symbols that have been assigned. That is done by category in the next section, and in "braille order" in Appendix G.
In this section we consider the assignments of symbols, all formed in accordance with the structural rules given in the preceding section, to specific meanings — graphic symbols (i.e. those that correspond to a sign that would appear physically in print) or indicators (i.e. those that do not directly correspond to concrete print signs, but that affect the meaning of nearby symbols in braille). The assignments are listed in groups according to the nature of the symbol.
Some indicators apply only to the immediately following symbol, while others initiate "modes" that extend over several symbols or even multiple words, until the mode is terminated. The mode of termination will vary with the indicator; it may be implicit (e.g. at a space) or explicit (that is, at a terminating indicator), but in any event the terminating condition should always be precisely defined.
Modes that persist indefinitely, that is until an explicit terminator is encountered, are sometimes referred to as "passages." In most instances, the extent of passages (other than passages in foreign languages or other non-UEB braille codes) should be confined to a section of text that is normally read as a unit — that is, a single paragraph, a single heading at any level, a single line of an outline, a stanza of a poem, or other comparable unit. (That last phrase is not perfectly precise, nor does it have to be; the intention is to limit "passages" to natural units of reasonable size, so that readers entering a text in the middle need not search too far back to be sure that all applicable indicators have been seen.) "Pages" and "lines" in the physical sense, that is those that are created simply as an accident of print formatting, are not natural units for the purpose of this definition.
Apart from the meanings that are limited to specific modes, such as the digits when in "numeric" mode, the meanings that are listed in this section are the basic (or "grade 1" or "uncontracted") meanings of the symbols. In those materials where contractions are used, that is in most materials, any symbols that can stand for contractions and that are located in such a way that the contraction meaning is allowed are to be read as contractions. For example, dots 2346 is listed here as meaning the mathematical "integral sign" (3.14), but in most contexts it would mean the letter-group "the". It would mean "integral sign" only when one of the grade 1 indicators (see section 3.2) applies, or in materials entirely in grade 1. Similarly, dots 235 would generally mean "ff" whenever it appeared in the designated context (between letters) in contracted materials; it would mean "exclamation mark" when not in that context and in any instance where a grade 1 indicator is applied to it.
In the braille edition of the listings, each braille symbol is preceded by a two-cell "dot locator," dots 46, 123456, that is not part of the symbol itself but is there to make the precise dot configuration clear, e.g. to distinguish dots 145 from the similar pattern dots 256.
The basic idea behind these indicators is that certain symbols
may have both a "grade
1 meaning" and one or more "contraction meanings" or a "numeric
meaning". Whenever the
grade 1 meaning of such a symbol occurs in a context such that it
could be misread as one of
the other meanings, then a grade 1 indicator is used to establish
the grade 1 meaning. For
example,
As in EB, dots 56 is often referred to as the "letter sign," even though its use in UEB is somewhat broader.
Grade 1 mode is established for the next symbol only by the single letter sign; up to the next space or grade 1 termination indicator, whichever comes first, by the double letter sign; and up to the grade 1 termination indicator by the triple letter sign.
Note that according to the more technical symbol construction
rules given in
Appendix B, the terminator is a sequence of two symbols, not a
single symbol; thus its use for
this kind of purpose is something of an exception. That use is
justified primarily by the
parallel with certain other terminations, such as for capitals,
and causes no conflict in
meaning because the apostrophe symbol
In general, numeric mode and other modes described below may be used within grade 1 mode. Likewise grade 1 mode may be used within other modes in any instance where a grade 2 or numeric meaning might otherwise be implied. Grade 1 is implicit immediately following numbers, as defined in more detail under "Numeric Indicators" below.
Grade 1 mode is the chief means by which braille symbols that can represent contractions can also be used for special symbols such as those needed in mathematics. It is a straightforward generalization of the familiar EB letter sign, with that concept both expanded and more precisely defined.
All of the assignments given throughout this section 3, except for the numeric meanings discussed under "Numeric Indicators" and the symbols defined for use only within special modes such as arrows, are grade 1 meanings.
The grade 1 indicator, also called the letter sign, should henceforth be used only for indicating grade 1 mode, and no longer for such EB uses as distinguishing the letter "a", "i" or "o" when used "as a letter" versus "as a word". Such distinctions are impossible to automate and have no interpretive value for the reader, since those letters cannot stand for contractions. Of course a letter sign should still precede the letters a through j, though not the other letters, immediately following a number.
The transcriber guidelines for use and non-use of extended (word- and passage-length) grade 1 modes are discussed in section 5.2.
Either of the two symbols that establish capitals mode may occur immediately before the first affected letter or accent modifier to a letter. All letters up until the termination condition are then understood to be capitalized.
The double capital symbol affects only the letters that follow. Its effect is terminated by the first symbol that is not a letter nor an accent or ligature modifier to a letter. The terminators thus include any indicators other than accent or ligature modifiers, including of course the capitals mode terminator, and punctuation marks such as apostrophe, hyphen, slash, etc.
The triple capital symbol affects all letters up to the next capitals mode terminator. The interpretation of any nonletter that may occur within a capitals passage is not affected.
Note that dot 6 does not appear here as a single-symbol
capital indicator, because
technically there is no such symbol. Rather, a single dot 6
before nonspace is regarded as the
prefix portion of a single symbol. Many such symbols stand for
capital letters, and some are
just ordinary symbols for which the concept of capitalization is
not applicable. The dash (
Apart from the addition of the passage indicator, these rules for the most part only confirm existing EB practice, except that the effect of a double capital indicator is not understood to continue through a hyphen. That is partly for simplicity, to avoid having to distinguish various kinds of punctuation marks; partly because the existing rules are somewhat contradictory in that respect (hyphen being commonly used as an informal equivalent to the capitals terminator); and lastly so that, when a hyphen between fully capitalized words is used to break a line, the capitals for the second part are already present and need not be introduced exceptionally.
Since a lower-h (dots 236) can mean either a question mark or an opening quote, and moreover can mean the contraction "his" in when located suitably in grade 2, it is necessary to specify exactly when each meaning is implied. The following rules cover all the cases, even unlikely ones:
1. In grade 2 context dots 236 has the meaning "his" if and only if it is both followed by a space and also preceded directly by a space or separated from the preceding space only by capitalization or typeform indicators, or any combination thereof.
2. In cases where the contraction "his" does not apply, dots 236 means an opening nonspecific quotation if and only if it immediately follows a space, hyphen or dash or otherwise satisfies the left context conditions for sequences or symbols "standing alone" as given in section 4.2 [that is, only symbol(s) as defined in list (1) of that definition intervene between the space, hyphen or dash and the dots 236].
3. In all other cases, dots 236 means a question mark.
Mostly the above rules simply go along with existing EB usage and the commonsense idea that dots 236 means "his" when it's more or less alone, means opening quote at the beginning of a word, and means question mark elsewhere — but they also give very precise definitions to those concepts and clarify unusual cases. For example, two lower-h's
In a similar way and for the same reasons, dots 6, 236 in grade 2 context would mean "His" if and only if it is both followed by a space and also preceded directly by a space or separated from the preceding space only by typeform indicators. In all other circumstances, it would be an opening single quotation mark.
The nonspecific quotes, that is those that are not distinguished as to whether they are "double" or "single" or "Italian", should be used for the predominant quotes in all instances where the specific form of quotation marks has no technical significance (that is, in the great majority of cases).
When non-specific quotes are used in a document, their use should include all instances of that form of quote that meet the criteria of the previous paragraph, e.g. both outer quotes and second-level inner quotes.
When practical, it is desirable to provide nonintrusive means by which the braille reader can determine the original form of quotes, even in nontechnical cases. Transcriber's notes, inclusion on special symbols pages, or any other such means of providing the information are encouraged, as permitted by the production context.
Despite the normal preference for the nonspecific form, the nonspecific opening quote symbol may be used only where it would not be read as a question mark or as the contraction "his", as detailed in the rules given above. When the location would cause the nonspecific opening quote to be read incorrectly, the specific "double" or "single" or "Italian" quote mark would be used.
"Nondirectional" quotes other than apostrophes, that is quotes without any slant or curl to convey "opening" or "closing", are to be used only in those relatively rare cases when such quotes are distinguished from directional ones (as in a discourse on typography), are otherwise clearly intended (as in an ASCII listing), or there is no way to infer directionality from context. Otherwise, directional quotes, as determined by context and according to the foregoing rules, are to be used.
More specific reasons for some of these assignments and rules are as follows:
The roots (dots 236 and 356) in the quotation group are obviously motivated by the current signs. The dot 6 prefix for single quotes is also current practice for the opening quote, and has a single dot suggesting the print sign. The two dots in the 45 prefix also suggests the print sign.
The question mark is also just the current English question mark, with dot 56 used for clarity when needed to assure that it cannot be confused with the nonspecific opening quote mark nor with the "his" contraction.
In the parentheses group, the 126 root is mainly to keep all these signs fairly light, and is consistent with British maths. The dot-5 prefix for round parentheses is mainly to keep the most common sign lightest of all; in fact it has only 4 dots, the same as the lower-g used in current EB. It also suggests a distinctive feature of the print sign, namely the bulge in the middle. Following that idea, the 46 prefix for square brackets suggests the tabs at the top and bottom of the print sign, and the 456 for curly brace the points at top, bottom, and middle of the print sign. Finally, this group is consistent with the assignments (dots 4, 126 ... 4, 345) for the angular brackets (which are also the "less-than" and "greater-than" symbols).
For practical application, though not incorporated into the rules, the following further guidelines on quotes and apostrophes summarize the prevailing thinking of the committee:
Many quotation marks in print are actually "nondirectional", that is they are typographically indistinguishable as to whether they are opening or closing. Material prepared on ordinary typewriters or simpler (especially older) ASCII-based word processors, for example, tend to have only nondirectional quotes. Furthermore, apostrophes and single quotes are usually the same print character in such material — which includes the vast majority of current computer programs. The committee's decision to retain a distinct apostrophe in braille, and the directionality of quotes, was based mostly on the desire to continue a long-standing tradition that has not been shown to be overly burdensome for transcribers, nor to lead to meaningful ambiguities for readers. To a lesser degree, the decision was also based on a sense that it could be moving backwards if we ceased making distinctions that we are already making in braille at the very time when, due to advancing desktop printing technology, those same distinctions are actually becoming more common in print — and could, conceivably, become technically meaningful in some cases.
Accordingly, in all normal cases, quotes would be rendered directional, and distinguished from apostrophes, in the same way that they are now, that is on the basis of usage, primarily positioning, with typography giving only secondary clues. For strange character sequences, such as might occur in computer programs, the following guidelines were offered: (1) When in doubt as to whether a mark is an apostrophe or single quotation mark, treat it as an apostrophe. (Rationale: If it's that strangely placed, it's probably not a quote, and marks such as mathematical "primes" are most naturally treated as apostrophes. [But note that true primes, when distinguished from apostrophes, are distinguished also in braille.]) (2) When in doubt as to whether a quotation mark is opening or closing, treat it as closing. (Rationale: That way, the reader is not psychologically conditioned to expect a closing quote. Also, most pseudo-quotes that really mean something else, such as the symbols for feet, inches, minutes and seconds, occur just after the associated numbers, positioned as closing quotes.)
Note that each of the twelve initiating ("set numeric") symbols is both a graphic and an indicator; that is, they not only stand for certain print symbols but also affect succeeding symbols. Specifically, they cause the braille symbols a through j to be interpreted as 1 through 0, a mode which persists through all such symbols and also the period/decimal point (dots 256), comma (dot 2), numeric separator spaces (i.e. the defined "space-digit" symbols), and a simple numeric fraction line (defined in later section 3.9) — that is, until some other symbol occurs, including the case of a letter a through j governed by a grade 1 indicator.
Strictly speaking, the period and comma are not included within the numeric mode unless they are actually followed by another numeric symbol, i.e. a digit or the numeric fraction line. This fine point has no practical effect in ordinary cases, but could enter into the treatment of certain indices.
The twenty symbols designated "in numeric mode only" are not general grade 1 assignments, but are limited to numeric mode.
If an extended grade 1 mode is in effect at the beginning of a number, then its effect is suspended in the number itself, and then resumed after the number. If grade 1 mode is not in effect at the beginning, then any of the initiating symbols establishes "grade 1 word" mode as well as numeric mode, which practically speaking means that contractions may not follow a number until after a space or a grade 1 termination (dots 56, 3). This is to reflect the fact that letters immediately following numbers are usually not words but rather designators typical of part numbers and such. (Of course, it is possible to terminate the grade 1 mode in those exceptional cases where it is desirable to express an attached English word in its ordinary form.)
As with capitals, these rules yield braille that is quite consistent with existing EB practice, except for the change of the decimal point to use the same symbol as the period, the fact that numeric mode does not continue after a hyphen, and the treatment of separating spaces as described below. The first of these two changes is motivated in part by the desire to use just one symbol where one is used in print, whereas in current EB, the BAUK codes use dot 2 to represent the decimal point and the BANA codes use dots 46 for the same purpose. Moreover, the dots 46 character, being a prefix, can no longer stand alone as a symbol except before a space. The second change, i.e. terminating numeric mode at a hyphen, parallels the similar change for double capitals, and is motivated by similar reasons.
The "spaced numeric mode indicator" does nothing more than
allow one or more
spaces to intervene between the numeric prefix and certain roots
that would normally follow
immediately, namely the digits
Taking that idea a step further, all instances of a the
symbols
The ten symbols of the form
The usage rule is: spaces should be represented in this way when and only when they are clearly used as separators within a single number, not for spaces between distinct numbers, such as in a list of numbers or numbers in distinct columns. For that purpose, a number generally considered as a single unit should be treated as such, even though it may comprise distinct parts, e.g. a single telephone number would be considered as one number, even though it includes country, city, and exchange codes as parts.
It would seem that, more simply, we could just say that dot 5, when occurring between digits within a number, represents a separating space in print. However, the assignments and rule are phrased as they are in order to assign meanings only to properly formed UEB symbols; dot 5 by itself, except before space, is not a complete symbol.
Without this special representation of numeric separator spaces, each such space would interrupt the numeric mode and then the number sign would need to be repeated after the space. The combination of the space and a new number sign was regarded as a stronger break than the mere space in the print, possibly suggesting to the reader that a new number is beginning, not just a continuation of the same number. This special representation avoids that problem.
The main disadvantage of a special representation of separator spaces is that, in the print to braille direction, it introduces the element of judgment as to whether a given space is being used as a separator within a number or is simply a regular space between different numbers. However, practical cases where this would be a real difficulty for a human transcriber were regarded as unlikely, and if there are any cases where real doubt would exist, the rule would simply call for treatment as an ordinary space. Along these lines, a further objection could be that computer programs are ill-equipped to make such judgments. When it comes to certain kinds of input, e.g. scanned input or simple-ASCII files where the spaces might not be distinguished, that objection has merit. However, in files prepared (or edited from scanned input) on typical word processors, separator spaces would most likely be entered as special "hard" (also called "non-breaking") spaces, so that the components of the number could not possibly be broken at a line ending. In that case, the discrimination would be quite easy for programs.
In summary, while the element of judgment is always something we would prefer to avoid, the burden in this case seemed light, and easily outweighed by the benefit to the reader.
1. A telephone number, 508 555 7549:
2. Decimal equivalent of 2 to the negative 17 power (as presented in "Handbook of Mathematical Functions", National Bureau of Standards 1964, p. 1016):
The symbols establishing typeforms follow a consistent prefix-root pattern, wherein the prefix designates the specific typeform and the root determines the extent. Most of the prefixes used have an historical association in English Braille or some other mnemonic basis, e.g.: dots 46 for italics from EB, 456 for underlining (or "right-hand letter l"), 45 for bold (a right-hand "b" — also has fewer dots than for underlining, which is less common).
Motivations for the specific extent assignments were: the dot 2 for word extent is the lightest symbol, dots 2356 for passage extent the heaviest, which is consistent with both perceived frequency of use and the size of the extent itself, and dots 23 for the single-symbol has a mnemonic basis as a "left-hand letter sign".
The "word" extent covers all symbols up to the next space, even if some characters that could not be meaningfully affected by the mode intervene. (For example, there is no such thing as an "italic period," but a period in the middle of an italic word would not terminate the effect of the italic-word indicator.)
The single-symbol indicators, when applied to a contraction symbol, are understood to affect only the first letter of the contraction.
The transcriber-assigned symbol sets are in a series that can be extended indefinitely, as needed, for any meaningful print distinctions not covered by the fixed assignments. (For example, colors might be used to flag various parts of speech in a grammar text.)
It bears restating that these indicators are not intended to be used necessarily wherever the corresponding typeform is used in print, but only where such usage conveys a meaningful distinction, such as to show emphasis, or the difference between computer input and output, or the class of a variable in mathematics. The "script" indicator, for example, would normally be used only when variation from some other prevailing letter form is intentional and meaningful, not just to show that the original text happened to be in script. Likewise, any typeform used simply as part of the formatting style, for example boldface used in all main headings, should normally be ignored in the transcription.
When a typeform extends over several paragraphs, it is preferable that the intermediate closing indicators (all but the last) be omitted, even though the opening indicator be repeated at the beginning of each new paragraph.
The basic rule, naturally, is that indicators should be placed so that no incorrect information is given — that is, so that any symbols that could be affected by the indicator are within the scope of the indicator if, and only if, they in fact have the typeform in question. However, since some typeforms are irrelevant to some symbols (e.g. the aforementioned "italic" and "period"), there is often some latitude in placement for the sake of readability. Among other factors, nesting of typeforms (i.e. closing them in reverse order to opening) is generally to be preferred, when possible given other requirements. For transcribers, more guidance on the optimal placement of typeform indicators is given in section 5.3.
Dots 36 is to be used for both minus sign and hyphen if they are not distinguished in print. However, when a minus sign is actually distinguished as such in print, it must be represented as dots 5, 36. This is because well-printed mathematics texts normally do show the symbols differently, and the distinction is helpful and in certain cases possibly necessary to understanding. (An example of such a case would be the expression "interest-rate minus inflation-rate", with "minus" shown in print by a distinctive minus sign.)
The plus sign was chosen to be a lower sign patterned after (though not identical to) the British and certain other maths codes, and so as not to vary with grade. The viability of a two-cell plus sign was questioned, but it was noted that the British two-cell plus sign remains well accepted after many years of use, and this symbol is actually lighter by one dot.
The simple level changes occur frequently in mathematics, so it was considered important that they be single-cell symbols when already in grade 1 mode, which would be the norm in complex mathematical expressions. When used in literary context, for example to show footnote references, a grade 1 symbol indicator (i.e. letter sign) would often be needed to avoid reading "en" or "in". The "shapes" of the lower-e and lower-i symbols suggested those choices for the level change symbols themselves.
The scope of any of the four level change indicators, that is the symbol(s) affected by it, is the next "item". An item is defined as any of the following groupings, according to the symbol that immediately follows the level change indicator:
(1) If the next symbol initiates simple numeric mode, the item is the entire number. That would include any interior decimal points, commas, separator spaces, or simple numeric fraction lines, but not "final" commas or periods.
(2) If the next symbol is an opening general fraction indicator (defined below), the item is the entire general fraction, through the closing fraction indicator.
(3) If the next symbol is an opening radical indicator (defined below), the item is the entire radical, through the closing radical indicator.
(4) If the next symbol initiates an arrow (defined below), the item is the entire arrow.
(5) If the next symbol initiates an arbitrary shape (defined below), the item is the entire shape.
(6a) If the next symbol is an opening round parenthesis, the item is the entire expression through the matching closing round parenthesis. Note that other round parentheses may occur interior to the item, in balanced pairs. Other kinds of parentheses or brackets may occur interior to the item and may or may not be in balanced pairs; only the round parentheses are important in determining the closing one that matches opening one at the beginning of the item.
(6b) If the next symbol is an opening square bracket, the item is defined analogously to (6a), but with square brackets instead of round parentheses.
(6c) If the next symbol is an opening curly brace, the item is defined analogously to (6a), but with curly braces instead of round parentheses.
(7) If the next symbol is a braille grouping opening indicator, the item is the entire group of symbols through the matching braille grouping closing indicator.
(8) If none of the foregoing apply, the item is simply the next individual symbol.
For transcribers, braille grouping indicators should be used to indicate the extent of an item [form (7) in the above list] only when one of the other forms does not apply, or in other words the opening symbol would not correctly identify which of the listed cases applies. For example, in the case where a "half-open interval" such as
The braille grouping signs are never used to replace existing graphic symbols such as brackets, but are used to create a group in braille not implied by (1) to (6) above.
It is implicit in the definition of an "item" that braille grouping indicators must be used for clarity whenever more than one indicator (e.g. superscript and over-bar) can apply to the same item, and it would otherwise be unclear which indicator is to be applied first to form a larger item. See section 3.11 for a more complete discussion of this topic.
Left-displaced indices, e.g. an expression written at the subscript level and before the base symbol, are handled simply by using the corresponding ordinary index expression prior to the base symbol. (Note: In general, such "left indices" would follow a space — otherwise they would be indistinguishable from right indices applied to the symbol on the left).
If indices occur within indices, they are to be expressed in the same manner as if they were at the base level. In other words, the notation may be "nested".
This approach to indices is similar to that taken in the most recent Spanish unified mathematics code (Ref. 91c), although symbols more consistent with other UEB assignments have been selected. It is also isomorphic to the general approach taken by mathematical typesetting and content descriptive languages that are important for print purposes, namely LaTeX and certain SGML Document Type Definitions (DTD's) (Refs. 90d, 89c). Finally, in using a "relative" rather than "absolute" method of dealing with multiple index levels, it is consistent with British maths.
1. (In grade 1 passage) x squared:
2. (In grade 2) x squared:
3. (In grade 1 passage) x squared plus y cubed equals z to the fourth power:
4. (In grade 1 passage) H sub 2 end-sub O (the familiar formula for water):
5. (In grade 2) H sub 2 end-sub O:
6. (In grade 1 passage) 6 space m to the negative 2 power:
7. (In grade 1 passage) x sub 1 end-sub squared equals y sub 2 end-sub cubed:
8. (In grade 2) an area of 6 m squared in total.:
9. (In grade 2) It travelled at 60 ft s to the negative 1 power.:
10. (A footnote reference in superscript position, in grade 2:) In Smith sup 56 end-sup we find ...:
11. (In either grade:) 4 x to the 1.5 power y to the .5 power:
12. (In either grade:) 4 b to the 1.5 power c to the .5 power:
13. (In a grade 1 passage:) e to the x squared plus y squared power
14. (In a grade 1 passage:) e to the (x sub i+1 end-sub to the p sub i end-sub power space + space y sub j+1 end-sub to the q sub j end-sub) power (Note: parentheses not literally in print)
15. (In a grade 1 passage:) The sum from i=1 to n of x sub i end-sub squared (with summation limits written directly below and above a capital sigma)
16. (In a grade 1 passage:) The sum from i=1 to n of x sub i end-sub squared (with summation limits written to lower right and upper right of a capital sigma)
A simple numeric fraction line symbol is used only for a simple numeric fraction, that is one whose numerator and denominator both contain only digits, decimal points, commas, or separator spaces — in other words, symbols (other than the fraction line itself) that continue a single numeric item. In such a case, a single numeric fraction line symbol may be used within the numeric mode, between the numerator and denominator, and continues the numeric mode so that the entire fraction is regarded as a single numeric item.
The numeric fraction line would be read as a line between vertically (or near-vertically) arranged numbers only, never as a general fraction line between larger expressions, which are treated below. The numeric fraction line is also not used where the print is expressed linearly, using an ordinary slash (oblique stroke) character; in that case, the braille simply uses the corresponding symbols as the print, and in the same sequence.
The "general" indicator symbols are to allow linear representation of fractions that are written in print with numerator over the denominator, and where either the numerator or denominator is not completely numeric as defined above.
The opening and closing general symbols are therefore technically indicators, since there is no corresponding graphic in print, and must be used in symmetrically balanced pairs. After the opening indicator, the numerator expression is written, then the general fraction-line symbol, then the denominator, and finally the closing indicator. Both numerator and denominator may be any kind of expression whatever, including fractions of either simple numeric or general type.
As with level changes, fractions were considered so common in math contexts that it would be important that these symbols be single-cell in grade 1 mode, while in general literary context the added letter sign was not considered an undue burden. The beginning and ending symbols are balanced symmetrically.
1. Two and one-half cups sugar:
2. The fraction whose numerator is two and one-half and whose denominator is x plus y:
An ordinary radical, that is one with a vinculum extending over the affected expression (the radicand), is expressed in braille by using the opening symbol before the radicand and the closing symbol after. The radicand itself may be any expression whatsoever, and may therefore have radicals as well as other mathematical structures. In other words, radicals may be "nested".
The radical index, if any, is signified by a superscript expression in the standard form, immediately following the opening radical symbol.
The symbol for a radical without vinculum is used as a simple graphic symbol, corresponding to any such symbol used in print.
1. The mn-th root of xy:
2a. The square root of four (written with vinculum over the four, as would be usual in modern usage):
2b. The square root symbol followed by four (with no vinculum, which would usually mean the square root of 4, especially in older books, but in some modern contexts could be just a sequence of symbols):
3. The familiar quadratic formula (x equals the fraction: minus b plus-or-minus the square root of b squared minus four a c end-root all over two a) :
4. r equals the square root of x squared plus y squared end-root:
5. q equals the cube root of x cubed space plus y cubed space plus z cubed:
6. The square root of the fraction: 783.2 times 6.547 over 0.4628 end-fraction end-root equals 105.3:
7. 81 to the three-quarters power equals left-paren the fourth root of 81 right-paren cubed equals left-paren the square root of the square root of 81 right-paren cubed equals left-paren the square root of 9 right-paren cubed equals three cubed equals 27:
8. The [square root of sixteen-ninths] root of 81 equals the four-thirds root of 81 equals 81 to the three-quarters power equals 27:
For all of these, the term "item" is as defined in section 3.8.
1. x with bar over it
2. the expression x plus y with bar over it
3. AB with right arrow above
4. AB with right arrow below
5. not-equals
Note on the definition of "item" and precedence: that no precedence has been defined between certain indicators that apply to the next, previous, or surrounding items. So, for example (in grade 1)
A future committee could choose to adopt precedence rules, or in other words a defined order in which the indicators involving "items" are to be applied. This would be similar to the way that, in mathematics, "a plus b times c" is generally understood to mean "a plus (b times c)", not "(a plus b) times c" — because multiplication is understood to take place before addition, except as explicitly determined by parentheses. If, for example, it were defined that the "bar" indicator has precedence over the "superscript" indicator, then it would be clear in our example that the bar was first applied to the y, thereby making a new item that was then an appropriate object of the superscript.
While such precedence rules might eventually be desirable, the committee has determined that it would be better, at least until more experience is gained, simply to require that braille grouping symbols always be used in such cases. In other words, in the above example, one or the other of
These indicators introduce a special "arrow" mode that provides a systematic method for representing nonlooping arrows other than those described in section 3.11 or that may be otherwise specially defined.
A nonlooping arrow may be regarded as a line or "shaft", with definite end-points, that does not cross or close upon itself and upon which one or more "tips" are superimposed. Tips may occur at either end of the shaft and/or along the shaft.
Such an arrow is to be treated as an enclosure, much like a general shape (see section 3.19), but using specific enclosing indicators (and interior symbol assignments) appropriate to the components of arrows. The opening symbols are those listed above. The arrow termination is one of eight symbols expressing the overall orientation of the arrow, as follows:
All these terminating symbols have three dots, arranged in a consistent pattern that best describes the overall arrow orientation, that is the absolute direction of motion (in two dimensions) if one were to proceed in a straight line towards the "head" of the arrow, starting at the other end (the "tail").
The head of the arrow is decided, if possible, by examining the direction of those arrow tips that have direction, that is those that have a concave and convex side, with respect to the arrow shaft. The direction of such tips is towards the convex side. (Note that the "direction" we are speaking of for this purpose is one-dimensional, i.e. along the shaft, so only two directions are possible.) The complete rules for deciding arrow direction are:
(1) If there are directional tips, and all lead in the same direction, the head is the end that lies in that direction.
(2) If there are no directional tips, but one end has a tip and the other does not, the end with the tip is the head.
(3) In all other cases, the head of the arrow is deemed to be the end at the right, or in the case of strictly vertical arrows, at the top.
Simple arrows: The simplest arrows are those with a straight shaft of medium length, and a single full, common barbed (hence directional) tip at the head. (Note that this definition implies that the tip points outward from the shaft; otherwise that end would be the "tail" of the arrow.) In such cases, only the arrow indicator and the direction (terminator) are given. This defines sixteen symbols, each comprising either of the two arrow indicators followed by any one of the eight terminators giving overall orientation. These sixteen include the following examples:
General nonlooping arrows: Other arrows in this class employ the same enclosures, but between them the tip(s) and shaft segment(s) are transcribed, using the symbols given below. These items are expressed in logical order, that is starting with the arrow tail and progressing towards the head, even if that runs counter to the physical order (as in the case of an arrow oriented towards the left). Certain elements are omitted, corresponding to these reader rules:
(1) If no tip is transcribed, it is understood that an ordinary full barbed tip occurs at the arrow head, and there is no other tip.
(2) If no shaft is transcribed, it is understood that the shaft is a straight line of medium length. In this case, if no tip is transcribed, rule (1) also applies; if one tip is transcribed, it is at the head; if two tips are transcribed, the first is at the tail and the second at the head.
Symbols for shafts (which may be elongated by repetition) are:
Length distinctions not made in the print need not be made in the braille; i.e. "medium" length would be the normal default.
Symbols for tips are:
Some example of arrows that could be constructed would be:
Notes: As with general shapes, this approach to arrows creates, between the enclosures, a distinct "mode" wherein symbols are understood differently. The available symbols for shaft segments and tips can easily be extended, provided that the symbols for shafts, tips, and terminators are always kept distinct. It would also be possible to extend this general method to arrows that can close upon themselves or cross, if that were later found to be desirable. Finally, this general approach does not preclude the possibility of assigning shorter specific symbols to particular arrows that occur frequently. For instance, it was noted that the common symbol for a reversible reaction in chemistry (a half-barbed left arrow over a half-barbed right arrow), while constructible as two general arrows vertically composed, would be too clumsy in that form when the frequency of the symbol is taken into account, and so a specific sign has been assigned (see section 3.18).
As in the case of the hyphen and minus sign, the prime and the apostrophe are distinguished in braille only when they are distinguished in the print.
The actual setting-out of arrays into columns was regarded to be a format issue, and therefore not the specific concern of Committee II, while graphic symbols for enclosing arrays are properly a code issue.
These are just the basic symbols plus a preceding dot 6 to convey the big/multi-line condition. These symbols are to be repeated on each line spanned by the print symbol, and vertically aligned, thus defining the overall vertical extent.
While a "line sign" (see section 3.25) can sometimes be useful for indicating new lines (rows), e.g. as an option for note taking, the symbol should not be used for general presentation of arrays, because then the point and purpose of the two-dimensional arrangement would largely be lost for the braille reader.
For common shape symbols as typically used in geometry and trigonometry, see section 3.19.
The committee also recommends setting up a "transcriber-assigned symbols" series to be used for symbols that are rarely encountered in general but that occur frequently in a particular text, such as made-up symbols or symbols that have come into regional usage.
Together with other print symbols presented in previous sections, some of these symbols complete the set of characters defined in the American Standard Code for Information Interchange (ASCII), which is probably the best-known long-established standard list of characters for computers. By providing symbols for all of these, we provide for essentially all that were covered by the BANA Computer Braille Code (Ref. 87a).
The "visible space" is to be used regardless of the print device used for such purpose (e.g. delta, underline with up-tick at each end, or slashed b), or where the transcriber determines that a space is significant, in the same sense that "countable spaces" are used in BANA Computer Braille Code. It was noted that such visible spaces could not be used within radicals, since the same symbol is the radical terminator, but the committee concluded that there was no actual requirement for "visible spaces" in that circumstance.
The continuation indicators are to be used when it is technically necessary to show that a print line is continued, and to represent precisely the sequence of symbols right through the braille line break; they are not to be used when there is no need for technical precision or where formatting (such as indented runovers in poetry) effectively conveys the runover information adequately.
When a space occurs at the point of a break to be shown by a continuation indicator, its presence is conveyed by using the two-cell continuation indicator.
The cursor indicator symbol is used in the case where it is necessary to show the position of a "cursor" (a special blinking or otherwise distinctive shape, used to identify the focal point of activity on a computer screen). The placement of the cursor indicator is on a separate braille line, immediately under the symbol to which it refers (specifically the final cell thereof, in the case of a multi-cell symbol). There is no actual conflict with the horizontal juxtaposition indicator (section 3.20), since in the case of the cursor indicator the symbols on both sides would always be spaces, and that would never be true when using the horizontal juxtaposition indicator.
The ordinary single line bond uses the same symbol as the dash (section 3.4), a single electron shown as a cross uses the same symbol as the "times" cross (section 3.7), and a single electron shown as a small circle uses the same symbol as the "hollow dot" (section 3.16).
While the line bond symbols are to be used for the horizontal bonds only, they are general symbols and may be used in structure formulae where other bonds, such as vertical bonds, are represented using line-drawing symbols (see section 3.26).
The equilibrium arrow is normally to be used only for "double harpoons", the upper arrow having a half-barb on the upper side at the right end pointing right and the lower arrow having a half-barb on the lower side at the left end pointing left, both arrows being of equal length and weight. In cases where a transcriber familiar with the notation is quite certain that some variant double-arrow form has been used consistently throughout the text to have the same meaning, and the precise form itself does not appear, the transcriber may use the equilibrium arrow in braille with a transcriber's note documenting the substitution and the form of the symbol in print.
The equilibrium arrows that signify a trend to the right or left may be used when either emphasis or length or both distinguishes one of the directions as dominant. In cases where a transcriber familiar with the notation is quite certain that some variant double-arrow form has been used consistently throughout the text to have the same meaning as one of the trending arrows, and the precise form itself does not appear, the transcriber may use the trending arrow in braille with a transcriber's note documenting the substitution and the form of the symbol in print.
Unusual, ad-hoc and iconic symbols are considered in two categories: arbitrary described symbols (shapes) and combination symbols (see section 3.20). Only relatively uncommon symbols are to be treated in either category. That is, any common symbol is to be assigned a regular braille symbol in its own right, even if it would be possible to use a "shape" or "combination" treatment, and when such an assignment has been made then naturally that assignment is to be used.
After the opening indicator, the next symbol determines the manner of termination. If that symbol is the opening general enclosure (i.e. braille grouping symbol, dots 126, introduced in section 3.8), then all following symbols, which must be validly formed UEB symbols, through the next closing general enclosure (dots 345), including any space or dots 156, are part of the shape description. If the symbol following the opening indicator is not an opening general enclosure, then all following symbols through the next shape terminator (dots 156), or up to (but not including) the next space, make up the shape description. (For transcribers, this implies that a shape from the standard list would require a shape terminator if unspaced on the right.)
In accordance with our general principles (see A.2 in Appendix A), the "descriptions" for shapes are to be language-independent if possible, although it is recognized that such independence is frequently impractical, i.e. shapes will commonly be described by English words. Because the terminator symbol (dots 345 or dots 156) cannot be used within the description, grade 1 is generally to be used within the description. In fact, a common pattern will be to use a short series of initials or a single word for the sake of brevity, especially when a particular "shape" must be used frequently within a given text.
The definitions of all such symbols must be available to the reader. Shapes in the "transcriber-assigned" category would be listed in preliminary notes, according to producer custom (e.g. in a "special symbols page"). Possible examples would be
(Note: The second of these examples was drawn from Knuth's "The TEXbook" [Ref. 90c], where a curvy-road-sign symbol is used as an icon to flag passages treating difficult or tricky material.)
Shapes not in the transcriber-assigned category, i.e. without the dot 4, are to be from an official list of those shapes that are more generally encountered — but still not so frequently that a specific regular symbol is justified. The list as given was purposely brief, covering only the family of basic regular figures (though that family is open-ended, as noted). A future standing committee is envisioned as adding to that list gradually, as it becomes evident that certain shapes are useful to standardize. In many cases, adding to the list would be a matter of removing the dot 4 from a transcriber assignment that had proved useful and popular.
For filled and shaded shapes, the distinctive opening indicator should be used, and the remainder of the shape symbol composed in the same manner as for a basic or transcriber-assigned shape.
These indicators allow multiple symbols to be combined into a new single symbol. Again, only relatively uncommon symbols are to be treated by these mechanisms (see section 3.19); any specifically assigned braille symbol would always have preference.
Each of these indicators signals a combining of the item just prior with the item immediately following it, where "item" is as defined in section 3.8.
"Horizontal juxtaposition" is to be invoked only when two symbols are written in close proximity and it is clear from the usage that a new single symbol, distinct from the elementary symbols considered in sequence, is intended. Otherwise, symbols written one after the other should simply be brailled accordingly.
Likewise, "vertical juxtaposition" is to be considered only when a new single symbol is formed, and should not be confused with indices directly above or below, nor with vertical arrangements such as in columns of matrices. It was noted that these distinctions could require judgment in some instances, but that this was unavoidable and unlikely to present real problems in practice.
Bars and arrows are not normally to be treated using this mechanism; see section 3.11.
Some examples, with grade 1 presumed, would be:
These all use the dots 45 prefix to suggest something "up", which of course is the usual case with accents. The actual positioning, whether over or under, is implicit in the symbol itself, which precedes the affected letter.
The "over following capital letter" forms are so that the single dot indicating a capital comes before the accent indication, in accordance with current EB practice, while still technically observing UEB symbol structure formalities. In cases like this, it is clearly easier (especially for teaching purposes) to think of the dot 6 as if it were a distinct indicator in its own right.
These accent marks are not intended for use in those cases where foreign language passages, as defined in the current English Braille code, are to be transcribed. That case is covered separately; see later section 3.23. Rather, they are to be used only in "anglicized" words (again, relying on the current definition) or in other cases where accented letters are used in essentially English or technical context.
These symbols are to be used for linguistic accents (that is, those that express in some sense how the affected letter is pronounced), and not for modifiers in mathematics, despite the similarity of print appearance in some cases. (For example, the second derivative of the variable "u", expressed as the letter with two dots above, is visually similar to a u with umlaut.) The committee reasoned that linguistic accents are basically different from mathematical modifiers both in common human understanding and, usually, even in computer file coding (reflecting, for example, the fact that mathematical modifiers may apply to expressions larger than a single letter).
The placement of the accent before the affected letter is to provide timely warning to the reader that the pronunciation of the next letter is affected by an accent (as is conveyed in current EB, though only in a nonspecific way, by the preceding dot 4).
Some specific reasons for the symbol root assignments are: (1) The middle-c of the dieresis/umlaut symbol suggests the horizontal two-dot shape. (2) The French "c with cedilla" sign is the cedilla symbol root. (3) The dots 16 grave symbol root suggests the print shape. (4) The French "i with circumflex" sign is the circumflex symbol root. (5) The Spanish nyay sign is the root for the tilde accent symbol. (6) The dots 34 acute symbol root suggests the print shape, symmetrically to the grave symbol root.
The committee further recommends specific assignments for certain other accent marks and modifiers beyond the Western Europe-oriented ones assigned above, namely:
It was noted that Committee 4 might wish to consider the macron and breve, since those are commonly used as diacritical marks to indicate the length of vowels.
Considering the number of possible accents and diacritical marks, it was recommended that at least two "transcriber-assigned" accents be set aside for use as needed in specific texts.
The Greek symbols are based on the international standard (as in Refs. 72a, 77a and 84a) rather than upon the slightly different modern Greek alphabet (Ref. 90a). This is to be consistent with the Nemeth 1972 code, and with the intended use of these symbols in English and technical context only.
As with the accent marks, these symbols are not intended for use in those cases where actual Greek language passages, meeting the definition of "foreign language" passages as defined for current EB usage, are to be transcribed. That case is covered in a separate topic below (see section 3.23). Rather, they are to be used only in cases such as the word "microsecond" (with the "micro" written as a Greek mu), the names of fraternities and sororities, Greek letters such as pi and theta used as constants and variables in mathematics, and any other cases where Greek letters are used in essentially English or technical context.
The formation of the Greek alphabet illustrates a process that can be applied to other foreign alphabets as needed. The committee did consider several such possibilities, but no compelling candidate was put forward. (The Hebrew letters alef and bet are used in some mathematics, but in practice no other letters of that alphabet. Some Cyrillic letters may be used, but rarely. Old German [Gothic] script may be used, but that can better be as a "font", i.e. special typeform.) However, respecting the need for a future committee to assign another complete alphabet, or the needs of a specific transcription, all symbols having dots 456 prefix and an alphabetic root were reserved as a group for assignment to an alternate non-Roman alphabet.
In EB currently, even in a generally English context, a foreign language used as such is treated differently from English-language text. In American practice, the braille transcription of foreign-language material uses symbols that are mostly the same as those used in the uncontracted braille defined by speakers of that language, though some indicators and punctuation marks continue to follow English custom. In British practice, the same general concept prevails, although more of the foreign braille symbols and rules, including some contractions, may be used. In either case, the braille is thus logically connected to the braille that would be found in literature published entirely in that foreign language, which obviously benefits both those who are learning the language and those who already know it.
The only drawback to this current treatment is that the switches back and forth between English and the foreign language (or languages) are made implicitly. That is, the human brain is assumed capable of sensing which language is being written; no explicit indicators are given (except to the extent that font changes or formatting clues may be present). In some cases, this may only reflect the situation in the print, where sometimes the words of a foreign language may be distinguishable from English words only by referring to the larger context of presentation. This situation seems workable enough for human readers, but at the current state of the art (and for the foreseeable future) will inhibit automated translation, at least from braille to print. A blind teacher of foreign languages, for example, would have to "assist" the translation program in order to derive correct print from a braille file containing mixed English and foreign-language material. That is because this treatment of foreign languages inevitably introduces formal ambiguities into the braille.
For the automation of print to braille, the situation is essentially the same, at least for text that is scanned into the computer or otherwise unmarked as to language. However, the committee noted that material prepared on computers with one of the more advanced markup languages, or even with some word processors if properly used, can be annotated so that language changes are definitively indicated. (For instance, a word processor program might provide such a facility so that the correct language dictionary can be used for spell-checking.) In that case, it is relatively straightforward to automate production of braille with English and other languages intermixed.
A joint session of Committees 2 and 4 considered these matters and decided to assign symbols so that material transcribed in foreign-language braille codes (or for that matter any braille code other than UEB itself) within a general UEB context could be explicitly marked with code switching indicators (see Ref. 2001a, which remains current except for one guideline, and upon which the remainder of this section is based). (The exceptional guideline is a subsequent recommendation by Committee 2 that non-UEB passages always be explicitly closed prior to opening a new one.)
These symbols indicate text that is transcribed in a braille code other than UEB, such as Music Braille, established codes