key (or advice for readers)^†

For the visual appearance of attributions or citations,
see “note style” (below on this same page).

• advertisements lacking explicit copyright phrases in old print media, and
• occasional individual characters from computer fonts,

• ‘ [ ’ and ‘ ] ’: Most narrowly speaking, they are the only characters named brackets in U.S. terminology[*], but are more commonly--albeit redundantly--called square brackets. In ASCII, they're available as characters 91 and 93, which are encoded in Web addresses as “%5B” and “%5D”. Generally, they “indicate that the enclosed words have been added the original text by someone other than the original author”[**].  As a specific case, they indicate “interpolations in direct quotations”[**].  When they appear in original text that's being quoted on this Web site, parentheses are substituted (it might be preferable, for uniformity, to underline each (square) bracket in the original text, a notation already used for ellipsis points; however, some browsers render this so poorly that the underline isn't distinct from the bottom stroke of each bracket). In epigraphy, (square) brackets enclose “certain or probable restorations of letters no longer legible”[‡].
• ‘ { ’ and ‘ } ’: Most narrowly speaking, they are named braces, and more commonly called curly brackets in U.S. terminology[*], but apparently also “hooked brackets” in British English [‡].  In ASCII, they're available as characters 123 and 125, which are encoded in Web addresses as “%7B” and “%7D”. In epigraphy, braces enclose “duplicated letters”[‡].
• ‘ 〈 ’ and ‘ 〉 ’: The only pair in this entry lacking a single-word name, they're called angle brackets. They're not available in ASCII, nor even in the 8-bit character sets  ISO 8859-1  or  windows-1252. But 3 paired variations are available in Unicode:
  • ‘ 〈 ’ and ‘ 〉 ’: the left-and-right-pointing angle brackets, at U+2329 and U+232A, in the ‘Miscellaneous Technical’ range (U+2300--23FF): These are the ones described on Alan Wood's Web Site[†]  as being descended from the Monotype “Symbol” font. According to classicist font-designer David J. Perry[††], this is the Unicode pair that should be used for epigraphic or linguistic purposes. The symbols appear in his “Cardo” font (albeit with very few of the other characters in that range). Surprisingly, in recent versions of the corresponding “The Unicode Standard”, e.g.: 6.0 (2010), they're labelled “Deprecated angle brackets” [emphasis added].
  • ‘ ⟨ ’ and ‘ ⟩ ’: the left-and-right mathematical angle brackets, at U+27E8 and U+27E9, in the ‘Miscellaneous Mathematical Symbols-A’ range (U+27C0--27EF): Perry emphatically rejects (i.e.: “Do not use”) these “which are for mathematics” (his words), but presents as his next item, without any dismissive comment, the mathematical left-and-right white square brackets at U+27E6 and U+27E7.[††]  Sigh. This webmaster seriously doubts that engineers, who are a significant part of the intended audience not only for characters in the range of the bulleted item above, but also this range, will feel compelled to dismiss this pair of brackets that the designers of Unicode decided to label as “mathematical”.
  • ‘ 〈 ’ and ‘ 〉 ’: the left-and-right CJK angle brackets, at U+3008 and U+3009, in the ‘CJK Symbols and Punctuation’ range (U+3000--303F): The “CJK” identifies these as “Chinese-Japanese-Korean” characters; so not only might the behavior defined for East-Asian characters produce unintended results in Latin-alphabet text, but Perry points out that they are “normally double-width”[††]  (or so they appear on Alan Wood's Web Site[†], although apparently not so in this bulleted item).
  Alas, none of the characters for angle bracket are in WGL4, so all of the characters for this punctuation mark will be incorrectly displayed by MS IE6, including all the viable characters (i.e.: the 1^st 2 pairs), even when the font set as the default for the browser contains a glyph for the character.[#]  Until there's a need on this Web site for quotes within quotes in French, German, Greek, or Russian, it might be possible--without excessive effort--to substitute the similar characters ‘ ‹ ’ and ‘ › ’, which are in WGL (albeit encoded as 139 and 155, placing them within the internationally nonstandard character-code range (128--159) of Microsoft's  windows-1252), are correctly displayed by MS IE6, and look as if they're lower-case or small angle brackets, although they're really the single left-and-right-pointing angle quotation marks.[##]  In epigraphy, angle brackets enclose “letters missing in error, and corrections of the stonecutter's mistakes”.[‡]
• ‘ ( ’ and ‘ ) ’: Normally called parentheses in U.S. terminology, sometimes abbreviated to parens[*], but in British English  “curved brackets”[**] and “round brackets”[‡].  In ASCII, they're available as characters 40 and 41, which are encoded in Web addresses as “%28” and “%29”. When they appear in direct quotations, “the words inside them were part of the original text” that's being quoted. In epigraphy, parentheses enclose “completed abbreviations, and in [this the cited] book, they are also used for words that are not present in the Greek text, but are needed to complete the sense of the English translation”. The latter is exemplified by inserting “(son of)” when a Greek personal name in the genitive case is used with that meaning[‡], akin--if not identical--to the Slavic patronymic, e.g.: the Russian composer whose name is transliterated as Pyotr Ilych Tchaikovsky, identifying him as a son of Ilya Tchaikovsky.

• upper-&-lower-case Latin letters (e.g.: modern Turkish) beyond those in the English-language default set Windows 1252 (commonly but mistakenly named ‘ANSI’);
• basic Greek (i.e.: not polytonic);
• Cyrillic for Russian and Ukrainian (depending on location, the latter might be more appropriate);
• mathematical and technical symbols (albeit not all) from the ‘Symbol’ font that's available on apple Mac and Windows;
• combining diacritical and punctuation marks; and
• tiny subset of arrows & triangles, adequate to point to the 4 compass directions.

• Latin epigraphic characters, including early or alternative forms for numerals, symbols for weights, measures, and currency, and various obscure Latin symbols;
• polytonic Greek, including symbols dedicated to dasia & psili (the latter 2 regrettably are spacing, i.e.: noncombining) (Greek Extended: U+1F00--1FFF);
• Hebrew;
• “almost all” of the characters from the recommendation of the Medieval Unicode Font Initiative;
• Indoärabic numerals as glyphs in Arabic (U+0660--0669), although not the rest of Arabic.

From want of attention to this, important events have sometimes been misquoted by a year. In illustration may be considered the death of Queen Elizabeth ⟨I⟩. This occurred in what was then styled in England 24 March 1602 ⟨O.S.⟩, being the last day of that year. In France and wherever the N.S. prevailed, this day was described as 3 April, 1603. To avoid all possible ambiguity such dates are frequently expressed in fractional form as 24 March/3 April, 1602/3. In our modern histories years are always given according to N.S., but dates are otherwise left as they were originally recorded. Thus Queen Elizabeth is said to have died 24 March, 1603.[**]

• Latin alphabet, transliterated. These are shown herein as are other foreign words, typically in italics.
• Greek alphabet. These are shown within guillemets: ‘ « ’ and ‘ » ’.[#]  Not only is that the custom in Greece for modern Greek quotations, but it also serves as a signal to the reader for occasional instances of text in Greek that just happened to use only those Greek letters that are easy to mistake for letters in the Latin alphabet. Displaying those Greek characters requires that a site visitor's browser support Unicode. See “character sets” (this page, above).

• dasia a.k.a. ‘rough breathing’ (Latin spiritus asper) mark (  ̔ ) or ( ῾ ). It's a curved mark facing in the same direction as ‘c’, but often shaped more like a mirrored comma (in fact, its initially recommended form in Unicode is assigned the generic name “combining reversed comma above”). Except for being a combining diacritical character (e.g.: ἁ  ἑ  ἡ  ἱ  ὁ  ὑ  ὡ ), it's easily confused with a typesetter's left single-quote. To allow room for a centered accent, the dasia customarily appears left-of-center over vowels (i.e.:  ά̔  έ̔  ή̔  ί̔  ό̔  ύ̔  ώ̔ ). Alas, the dasia is not always rendered well in older browsers, and its combination with other accents even worse, perhaps the result of programming neglect that might've been rationalized by its being abolished in the Greek official modernization. But it's especially important for the ‘dasia’ to be recognizable[##], because it signifies the sound rendered in English by ‘h’, so it's shown on this Web-site whenever a source provides it. It may appear in some transliterations as a leading ‘h’, possibly superscripted ( ^h ).
• psili a.k.a. ‘smooth breathing’ (Latin spiritus lenis) mark (  ̓ ) or ( ᾿ ). It's a curved mark facing in the same direction as a comma, which it's often shaped like. It's easily confused with a typesetter's right single-quote. It was also abolished by the Greek modernization, and is not used herein except in this bulleted item.
• oxía ( ´ ): It's practically the same mark as the right-leaning diagonal European acute accent. It's really the only diacritical mark that survived the Greek modernization (but see the interloping tónos, below).
• varía ( ` ): It's practically the same mark as the left-leaning diagonal European grave accent. It was also abolished by the Greek modernization, and is not used herein except in this bulleted item.
• tónos ( ΄ ), or the universally available approximation for displaying its shape, the apostrophe ( ' ): Although the word itself is Greek, meaning “accent”, the mark has a form never used as an accent by native Greeks, according to a native Greek typographer: It's an unslanted downward-pointing wedge mark, much like the unslanted straight form of the ASCII apostrophe ( ' ), centered over a Greek miniscule letter (i.e.:  ά  έ  ή  ί  ό  ύ  ώ ), instead of being off to either side, but preceding a Greek majuscule letter. Its name & form is reportedly an invention of the Unicode Consortium, for the only diacritical mark that survived the Greek modernization: the oxía. So, interpret it as an oxía whenever a browser displays it in this form.
• perispoméni: A problematic diacritical mark, “historically it started as a combination of acute and grave”, but in modern fonts, it doesn't appear as the modern circumflex ( ^ ) that a sensible reader might expect. Instead, it appears in 3 different forms, i.e.: the marks more familiar as the tilde, macron, or inverted breve, of which only 1 is used for any specific font. Alas, “in some typefaces, the breathing is connected to the perispoméni so that they seem to form a single accent” (e.g.: a macron over a tónos, whose combination looks like a tiny ‘ t ’). Being another accent, a perispoméni is shown on this Web-site whenever a source provides it. It was also abolished by the Greek modernization, so it warrants no further discussion at this time,
• Other obscure Greek diacritics that won't appear herein, and thus warrant not even a mention at this time.

• There were no miniscule letters in Classical Latin, which is the Latin written while Jesus the Christ and His Apostles walked the Earth.
• The italics that are conventional in English-language typography have many uses, so it's already difficult enough for readers to distinguish the significance of each use.

• adjectives: their 1--3 nominative singular forms; when there are (exactly) 3, they identify the endings for each of the 3 grammatical genders.
• adverbs: each has only 1 form--always.
• nouns: their nominative singular and genitive singular forms, which together imply the declension to which each word belongs.
• verbs: their 4 (sometimes fewer) principal parts, which together imply the conjugation to which each word belongs.

• “naïve”, where ‘ ï ’ indicates the word's pronounciation as “na_h-eev' ” instead of “na_yv”.
• “coöperate”, a compound where ‘ ö ’ indicates the word's pronounciation as, i.a., “co_h-op' -u_h-ra_yt” instead of “coop' -u_h-ra_yt” (the latter erroneously combining the first 2 syllables into the word for a poultry domicile that rhymes with “loop”) [†]. Compare it to a word whose common spelling differs only in its penultimate letter, but it should never bear a diëresis: “cooperage”, its correct pronunciations, i.a., “coop' -u_hr-ij” or “coo' -pu_h-rij”.

• minimizing the insertion of middots even if doing so obscures the inflected endings of the word, e.g.: “vesp ·er, -eris | -eri ” (despite “-er-” not being part of its inflected ending); or
• accurately indicating the inflected endings of the word, e.g.: “vesper·, -is | -i ” (where the middot placed at the end signifies that its nominative case has no inflected ending at all).

For the principles that justify inclusion of notes as attributions or citations,
see “attributions or citations” (above on this same page).

• author name(s): Reversing the name of the first author in a list, so that it appears with the surname first (while presenting the names of any coäuthors in their natural order).
• publication place: Preceding the name of the publisher, even though the place-of-publication is largely irrelevant nowadays: For many decades in the U.S.A., that place has been “New York City”, and when overlooked, it's still the default place. But to order a book (or to evaluate it without having it in one's own hands, e.g., by description & comments on a Web site), the name of its publisher is much more important. Especially back when certain publishers (e.g.: TAN Books) could be relied on to produce new reprints of fully traditional Catholic books without “updating” them to the theology, ecclesiology, and practices of the Novus Ordo religion of Vatican II.
• “Superior figures-- small figures set above the line--are generally used as references to footnotes. The footnote itself is identified in most cases by ⟨prefixing with⟩ a corresponding superior figure. [....] it is best to number them consecutively throughout a chapter or article.”[**]  Ah, yes. In the earliest years of this Web site, even its longest pages did seem to have a length not much longer than a chapter of a published textbook. Alas, even the relatively recent update of the computer language for Web-page mark-up (i.e.: HTML) still provides no way for Web browsers to identify the logical connection between any specific definition or note, and reference(s) to either, unlike word-processors for personal computers that have been performing that function for more than 3 decades on far less capable computers [##]. And meanwhile, the longest pages on this Web site have grown much longer.
  
  There is a practical alternative: “Reference marks may be used instead of superior figures. If several marks are used on a page the proper sequence is”[**], simply listed:  ‘*’, ‘†’, ‘‡’, ‘§’, ‘∥’ (the inelegant ASCII digraph ‘||’ might suffice as substitute), ‘¶’.  All well & good, perhaps. But on this Web site, various special characters will be substituted [###]:
  • drawn from ASCII, e.g.: ‘#’, ‘$’, ‘*’, ‘+’, ‘@’;
  • drawn from the Web-server & Western-European-language default character set ISO-8859-1, e.g.: ‘¢’ ‘×’, ‘÷’; and
  • drawn from Unicode, and displayed via their internationally standard symbolic or numeric identification as devised for Web pages, e.g.: ‘†’, ‘‡’, ‘∆’ (or‘Δ’), ‘∇’, ‘√’, ‘◊’.
  Sometimes the characters will be doubled or tripled (as of this update early in 2016, none have yet been quadrupled), but they will always be enclosed by brackets, e.g.: “[***]”.  There doesn't seem to be any practical alternative that avoids the hassles of maintaining numerical order, especially the renumbering required by editing rules when new notes are inserted [#]. It's a fine example of an approach that, despite its limitations, seemed to be the most satisfactory solution at the time, and still does.
  
  Neither superior figures nor reference marks address the reference problems created by the Internet custom that a revised Web page can replace an existing Web page whenever any of its webmaster(s) &al. decide(s) that such a change is appropriate. But the Internet has no established mechanism to maintain synchronization between a Web site that originates information, and a Web-site elsewhere that references that same information. This situation is the basis of broken links (for which Wikipedia, e.g., is notorious among people accustomed to trying to track down sources for claims in its articles), whether resulting from a Web site being abandoned, reörganizing its file-structure, or renaming its files[#]. Revisions in printed books are much less problematic, because physical considerations in their production confine changes in content to editions and printings that can be cited and verified.

[...] representing fairly, proportionately, and as far as possible, without bias, all of the significant views that have been published by reliable sources on a topic. [....] This policy is nonnegotiable [....]

• link to a specific entry on the page (Web address contains '#'): Move the cursor to the browser's Web-address input area (usually appearing as a cut-out with a white-background): Select (i.e.: highlight) the '#' and everything to the right of it, and press the 'Delete' (or 'Backspace') key. Continue with the instructions at the 2^nd bulleted item.
• link to entire page (Web address contains no '#'): Append the following immediately after (i.e.: without any spaces separating it from) the Web address in the browser's Web-address input area:

  ?print

• “gloss.sht”: 154 entries occupying 710 KB, with a 90 KB line-illustration displayed at the top, and 35 KB of small images in the text, yielding a default print-preview totalling 96 pp.; and
• “notion.sht”: 130 entries occupying 576 KB, with a 100 KB line-illustration displayed at the top, and 70 KB of small images in the text, yielding a default print-preview totalling 77 pp.

• An authoritative source: This webmaster recently became the grateful recipient of a 20^th-Century Hebrew-English Lexicon, focused on the Old Testament, and organized alphabetically--or is that abjadically?--first by roots, then by their derivatives.
• Greatest availability of characters: Unicode has encoded a Hebrew character set, including diacritical marks, since version 1.0.0 (Oct. 1991), at U+0590--05FF. By contrast, the historically more proper Phoenician was not encoded until version 5.0 (July 2006), and exiled to U+10900--1091F, up in the Supplementary Multilingual Plane. Up there, it would have the comparably proper Imperial Aramaic for company, at U+10840--1085F, but it was not encoded until version 5.2 (Oct. 2009), and fonts that support it are a wee bit too scarce to expect readers to be able to display the characters.
• Unreliable diacritical marks: As of the 2^nd decade of the 21^st Century, correct placement of diacritical marks for Hebrew seems beyond the capabilities of Web browsers--or at least of many versions of them--still in common use. For Hebrew, they're not the largish marks of the Western European languages that use the Latin alphabet; instead, they can be eye-straining dots of scarce-few pixels, which need to be displayed on a particular side of, or within, a particular letter. From examination of the “notions” entry for “Messias”, it seems that marks are misplaced when a single letter requires more than a single (type of) mark.

Arabic, which is rich in earthy phrases, loaned some of them to Hebrew. If you get into a traffic accident with a Tel Aviv cab driver you will probably hear some of them. Ben Yehuda ⟨the “father of modern Hebrew”⟩ would not have approved but it is inevitable.

• ‘ א ’ (aleph) transliterated to the glottal plosive ‘ ʔ ’, and
• ‘ ע ’ (ain) transliterated to the voiced pharyngeal fricative ‘ ʕ ’.