{
    "paper_id": "W89-0128",
    "header": {
        "generated_with": "S2ORC 1.0.0",
        "date_generated": "2023-01-19T03:45:29.599944Z"
    },
    "title": "Database Systems for Lexicographic Work",
    "authors": [
        {
            "first": "P",
            "middle": [
                "\u00bb"
            ],
            "last": "Bj\u00f6rn",
            "suffix": "",
            "affiliation": {},
            "email": ""
        },
        {
            "first": "J\u00f6rgen",
            "middle": [
                "F"
            ],
            "last": "Svavarsson",
            "suffix": "",
            "affiliation": {},
            "email": ""
        },
        {
            "first": "",
            "middle": [],
            "last": "Ind",
            "suffix": "",
            "affiliation": {},
            "email": ""
        }
    ],
    "year": "",
    "venue": null,
    "identifiers": {},
    "abstract": "At the Institute o f Lexicography of the University of Iceland work revolves around very large collections of data in computers. There are mainly two types of databases which are kept in computer storage. The first one is a relatively simple database containing the whole vo cabulary o f the main collection o f the Institute, along with the age, number o f citation, word class, word type, and oldest citation for each word. For maintaining this database there is no need for a very complex or powerful database system, but it must be fast since it contains over 600,000 words. The second datab^tse is much more complex. It contains the lexico graphic analysis and is used to construct the dictionary itself. A system like this must be more \"intelligent\" and more flexible than the first one, but speed is not as important a feature. This paper describes some of the properties o f the database systems we have been working on under MS-DOS and UNIX at the Institute. 1 The Background T h e In stitu te o f L e x ic o g r a p h y a t th e U n iv e rs ity o f Ice la n d w as esta b lish ed in 1 9 4 7 w ith th e m a jo r p u r p o s e o f m a ld n g a h is to rica l d ic tio n a r y o f th e Icela n d ic la n g u a g e , co v e r in g th e p e r io d 1540 t o th e presen t. O v e r th e pa st fo u r d eca d e s, m a jo r e ffo rt h as b e e n p u t in to th e e x c e r p tio n , b u ild in g u p a c o lle c tio n o f so m e 2 .6 m illion c ita tio n s. T h e s ta te o f e x c e r p tio n is n ow su ch th a t it is ' c o m p le te ' fo r th e 1 6th , 1 7th a n d 18 th cen tu ries, 'fa irly c o m p le t e ' fo r th e 19th cen tu ry, and ' in c o m p le te b u t s u b s ta n tia l' fo r th e 2 0 th cen tu ry. T h e c o lle c tio n e n com p a sses a v o c a b u la r y in e x ce s s o f 6 0 0 ,0 0 0 w ord s. W h e n c o n fr o n te d w ith su ch a m a ssiv e a m o u n t o f m a te ria l, th e q u e stio n arises as t o w h a t w o u ld b e th e n a tu ra l w a y o f p r o c e e d in g w ith th e e d itin g. T h e first q u e s tio n o n e w o u ld p r e s u m a b ly a sk is: H o w has it b e e n d o n e ? W ell, th e answ er t o th a t q u e s tio n is p r e tty w ell k n o w n. T ra d itio n a lly , a fte r th e m a te ria l has been c o lle c te d , o r a t lea st a s u b sta n tia l p a rt o f it, th e e d it o r sits d o w n w ith th e first fe w h u n d re d slip s fr o m th e first b o x a n d tries n o t t o th in k t o o m u ch a b o u t th e",
    "pdf_parse": {
        "paper_id": "W89-0128",
        "_pdf_hash": "",
        "abstract": [
            {
                "text": "At the Institute o f Lexicography of the University of Iceland work revolves around very large collections of data in computers. There are mainly two types of databases which are kept in computer storage. The first one is a relatively simple database containing the whole vo cabulary o f the main collection o f the Institute, along with the age, number o f citation, word class, word type, and oldest citation for each word. For maintaining this database there is no need for a very complex or powerful database system, but it must be fast since it contains over 600,000 words. The second datab^tse is much more complex. It contains the lexico graphic analysis and is used to construct the dictionary itself. A system like this must be more \"intelligent\" and more flexible than the first one, but speed is not as important a feature. This paper describes some of the properties o f the database systems we have been working on under MS-DOS and UNIX at the Institute. 1 The Background T h e In stitu te o f L e x ic o g r a p h y a t th e U n iv e rs ity o f Ice la n d w as esta b lish ed in 1 9 4 7 w ith th e m a jo r p u r p o s e o f m a ld n g a h is to rica l d ic tio n a r y o f th e Icela n d ic la n g u a g e , co v e r in g th e p e r io d 1540 t o th e presen t. O v e r th e pa st fo u r d eca d e s, m a jo r e ffo rt h as b e e n p u t in to th e e x c e r p tio n , b u ild in g u p a c o lle c tio n o f so m e 2 .6 m illion c ita tio n s. T h e s ta te o f e x c e r p tio n is n ow su ch th a t it is ' c o m p le te ' fo r th e 1 6th , 1 7th a n d 18 th cen tu ries, 'fa irly c o m p le t e ' fo r th e 19th cen tu ry, and ' in c o m p le te b u t s u b s ta n tia l' fo r th e 2 0 th cen tu ry. T h e c o lle c tio n e n com p a sses a v o c a b u la r y in e x ce s s o f 6 0 0 ,0 0 0 w ord s. W h e n c o n fr o n te d w ith su ch a m a ssiv e a m o u n t o f m a te ria l, th e q u e stio n arises as t o w h a t w o u ld b e th e n a tu ra l w a y o f p r o c e e d in g w ith th e e d itin g. T h e first q u e s tio n o n e w o u ld p r e s u m a b ly a sk is: H o w has it b e e n d o n e ? W ell, th e answ er t o th a t q u e s tio n is p r e tty w ell k n o w n. T ra d itio n a lly , a fte r th e m a te ria l has been c o lle c te d , o r a t lea st a s u b sta n tia l p a rt o f it, th e e d it o r sits d o w n w ith th e first fe w h u n d re d slip s fr o m th e first b o x a n d tries n o t t o th in k t o o m u ch a b o u t th e",
                "cite_spans": [],
                "ref_spans": [],
                "eq_spans": [],
                "section": "Abstract",
                "sec_num": null
            }
        ],
        "body_text": [
            {
                "text": "e n ly a n d co n sta n tly , a n y rev ision s, w h ich m a y b e ca lled fo r a t a la te r d a te , m a y lea d t o th e n ecessity o f re d o in g th e a n a lysis, ta k in g the m a n u s crip t a p a r t, as it w ere, a n d p u ttin g it b a ck to g e th e r again . There were, of course, other reasons as well. Unix is well-known for its outstanding collec tion of tools, its preeminence in dealing with text files, the easy etccess to e-mail, etc.",
                "cite_spans": [],
                "ref_spans": [],
                "eq_spans": [],
                "section": "",
                "sec_num": null
            },
            {
                "text": "When compared with the description of relational database systems given, for example, by Date (1986) .",
                "cite_spans": [
                    {
                        "start": 89,
                        "end": 100,
                        "text": "Date (1986)",
                        "ref_id": "BIBREF3"
                    }
                ],
                "ref_spans": [],
                "eq_spans": [],
                "section": "T h e C om puter Databaise",
                "sec_num": "3.2"
            },
            {
                "text": "In textform it is about 30 Mb. ",
                "cite_spans": [],
                "ref_spans": [],
                "eq_spans": [],
                "section": "T h e C om puter Databaise",
                "sec_num": "3.2"
            },
            {
                "text": "2 \u2022 It h as a p ow e rfu l q u e ry la n g u a g e, sim ilar t o S Q L .",
                "cite_spans": [],
                "ref_spans": [],
                "eq_spans": [],
                "section": "UNIX F ig u re 4: C om p u terized tasks a t th e In s titu te , as divided between M S -D O S an d U nix a t present.",
                "sec_num": null
            },
            {
                "text": "T h e m a jo r d isa d v a n ta g es o f A d v a n c e d R e v e la tio n are:",
                "cite_spans": [],
                "ref_spans": [],
                "eq_spans": [],
                "section": "UNIX F ig u re 4: C om p u terized tasks a t th e In s titu te , as divided between M S -D O S an d U nix a t present.",
                "sec_num": null
            },
            {
                "text": "\u2022 It is t o o slow . T h e c ita tio n s fo r e a ch w o r d a re k ep t in sep a ra te files w h ich a lso co n ta in the a n a ly sis. J 6 n H ilm a r J d n sson , in his p a p e r , d e scrib e s th e fields w h ich are en tered in to th ese files a n T w o scree n in p u t fo r m s are u sed fo r th e e d ito r ia l a n a ly sis, o n e fo r th e form a l a n a ly sis , th e o th e r fo r th e s e m a n tic a n a lysis. W h e n th e e d it o r sta rts on a p a rtic u la r p h a se o f th e a n a ly sis, h e o r she ca n use th e q u e ry m ech a n ism to arra n ge th e c it a tio n s in a n ",
                "cite_spans": [],
                "ref_spans": [],
                "eq_spans": [],
                "section": "UNIX F ig u re 4: C om p u terized tasks a t th e In s titu te , as divided between M S -D O S an d U nix a t present.",
                "sec_num": null
            },
            {
                "text": "Proceedings of NODALIDA 1989",
                "cite_spans": [],
                "ref_spans": [],
                "eq_spans": [],
                "section": "",
                "sec_num": null
            },
            {
                "text": "Svavaxsson & J\u00f6rgen Pind: Databases for Lexicography",
                "cite_spans": [],
                "ref_spans": [],
                "eq_spans": [],
                "section": "",
                "sec_num": null
            },
            {
                "text": "ComputationalLinguistics -Reykjavik 1989",
                "cite_spans": [],
                "ref_spans": [],
                "eq_spans": [],
                "section": "",
                "sec_num": null
            },
            {
                "text": "Svavarsson & J\u00f6rgen Pind: Databases for Lexicography",
                "cite_spans": [],
                "ref_spans": [],
                "eq_spans": [],
                "section": "",
                "sec_num": null
            },
            {
                "text": "U n ix so it is n o t p o s s ib le fo r us t o e la b o r a te in d e ta il as t o h o w th e p r o g r a m s w ill b e im p le m e n te d , b u t fig u re 7 sh ow s o u r p resen t id ea s.",
                "cite_spans": [],
                "ref_spans": [],
                "eq_spans": [],
                "section": "",
                "sec_num": null
            }
        ],
        "back_matter": [],
        "bib_entries": {
            "BIBREF0": {
                "ref_id": "b0",
                "title": "Um stofnhlutagreiningu samsettra orda. BA thesis in prepa ration",
                "authors": [
                    {
                        "first": "Kristin",
                        "middle": [],
                        "last": "Bjaxnad\u00e9ttir",
                        "suffix": ""
                    }
                ],
                "year": 1990,
                "venue": "",
                "volume": "",
                "issue": "",
                "pages": "",
                "other_ids": {},
                "num": null,
                "urls": [],
                "raw_text": "Kristin Bjaxnad\u00e9ttir. 1990. Um stofnhlutagreiningu samsettra orda. BA thesis in prepa ration.",
                "links": null
            },
            "BIBREF1": {
                "ref_id": "b1",
                "title": "Sigfus Bl\u00f6ndal. 1920-1924. fslensk-d\u00f6nsk or\u00e9ab\u00e5k",
                "authors": [],
                "year": null,
                "venue": "",
                "volume": "",
                "issue": "",
                "pages": "",
                "other_ids": {},
                "num": null,
                "urls": [],
                "raw_text": "Sigfus Bl\u00f6ndal. 1920-1924. fslensk-d\u00f6nsk or\u00e9ab\u00e5k. Reykjavik.",
                "links": null
            },
            "BIBREF2": {
                "ref_id": "b2",
                "title": "Advanced Revelation-Documentation, 4 volumes",
                "authors": [],
                "year": 1987,
                "venue": "",
                "volume": "",
                "issue": "",
                "pages": "",
                "other_ids": {},
                "num": null,
                "urls": [],
                "raw_text": "Cosmos. 1987. Advanced Revelation-Documentation, 4 volumes. Cosmos, Seattle, Washington.",
                "links": null
            },
            "BIBREF3": {
                "ref_id": "b3",
                "title": "An Introduction to Database Systems",
                "authors": [
                    {
                        "first": "C",
                        "middle": [
                            "J"
                        ],
                        "last": "Date",
                        "suffix": ""
                    }
                ],
                "year": 1986,
                "venue": "",
                "volume": "",
                "issue": "",
                "pages": "",
                "other_ids": {},
                "num": null,
                "urls": [],
                "raw_text": "Date, C. J. 1986. An Introduction to Database Systems. Addison-Wesley, Reading, Mass.",
                "links": null
            },
            "BIBREF4": {
                "ref_id": "b4",
                "title": "A Standardized Dictionary o f Icelandic Verbs",
                "authors": [
                    {
                        "first": "Jon",
                        "middle": [],
                        "last": "Hilmar",
                        "suffix": ""
                    },
                    {
                        "first": "J\u00f6nsson",
                        "middle": [],
                        "last": "",
                        "suffix": ""
                    }
                ],
                "year": 1989,
                "venue": "",
                "volume": "",
                "issue": "",
                "pages": "",
                "other_ids": {},
                "num": null,
                "urls": [],
                "raw_text": "Jon Hilmar J\u00f6nsson. 1989. A Standardized Dictionary o f Icelandic Verbs. (This vol ume).",
                "links": null
            },
            "BIBREF5": {
                "ref_id": "b5",
                "title": "Computers, Typesetting, and Lexicography",
                "authors": [
                    {
                        "first": "J\u00f6rgen",
                        "middle": [],
                        "last": "Find",
                        "suffix": ""
                    }
                ],
                "year": 1989,
                "venue": "",
                "volume": "",
                "issue": "",
                "pages": "",
                "other_ids": {},
                "num": null,
                "urls": [],
                "raw_text": "J\u00f6rgen Find. 1989. Computers, Typesetting, and Lexicography. (This volume).",
                "links": null
            },
            "BIBREF6": {
                "ref_id": "b6",
                "title": "On the Making o f the Middle English Dictionary. Dictionaries: Journal o f the Dictionary Society o f",
                "authors": [
                    {
                        "first": "Sherman",
                        "middle": [],
                        "last": "Kuhn",
                        "suffix": ""
                    }
                ],
                "year": 1982,
                "venue": "North America",
                "volume": "4",
                "issue": "",
                "pages": "14--41",
                "other_ids": {},
                "num": null,
                "urls": [],
                "raw_text": "Kuhn, Sherman. 1982. On the Making o f the Middle English Dictionary. Dictionaries: Journal o f the Dictionary Society o f North America 4:14-41.",
                "links": null
            },
            "BIBREF7": {
                "ref_id": "b7",
                "title": "Hypertext and the Oxford English Dictionary",
                "authors": [
                    {
                        "first": "Darrel",
                        "middle": [
                            "R"
                        ],
                        "last": "Raymond",
                        "suffix": ""
                    },
                    {
                        "first": "",
                        "middle": [],
                        "last": "Frank W M . Tompa",
                        "suffix": ""
                    }
                ],
                "year": 1986,
                "venue": "",
                "volume": "31",
                "issue": "",
                "pages": "871--879",
                "other_ids": {},
                "num": null,
                "urls": [],
                "raw_text": "Raymond, Darrel R., and Frank W m . Tompa. 1986. Hypertext and the Oxford English Dictionary. Communications o f the A C M 31:871-879.",
                "links": null
            },
            "BIBREF8": {
                "ref_id": "b8",
                "title": "Pick, Coherent and Theos",
                "authors": [
                    {
                        "first": "Marc",
                        "middle": [
                            "J"
                        ],
                        "last": "Rochkind",
                        "suffix": ""
                    }
                ],
                "year": 1986,
                "venue": "Byte",
                "volume": "10",
                "issue": "",
                "pages": "231--239",
                "other_ids": {},
                "num": null,
                "urls": [],
                "raw_text": "Rochkind, Marc. J. 1986. Pick, Coherent and Theos. Byte 10(ll):231-239.",
                "links": null
            },
            "BIBREF9": {
                "ref_id": "b9",
                "title": "Creating a Lexical Database from a Dictionary",
                "authors": [
                    {
                        "first": "Rudolf",
                        "middle": [],
                        "last": "Rydstedt",
                        "suffix": ""
                    }
                ],
                "year": 1988,
                "venue": "Studies in Computer-Aided Lexicology. Data Linguistica, 18. Almqvist & Wiksell",
                "volume": "",
                "issue": "",
                "pages": "",
                "other_ids": {},
                "num": null,
                "urls": [],
                "raw_text": "Rydstedt, Rudolf. 1988. Creating a Lexical Database from a Dictionary. Martin Geller stam [ed.]. Studies in Computer-Aided Lexicology. Data Linguistica, 18. Almqvist & Wiksell, Stockholm.",
                "links": null
            }
        },
        "ref_entries": {
            "FIGREF0": {
                "type_str": "figure",
                "uris": null,
                "num": null,
                "text": "The procedure o f making o f historical dictionary 2.6 m illion slips aw aitin g! A s m a teria l g a th ers, it is b r o u g h t o u t in in sta llm en ts o v e r a p e r io d o f d eca d e s, so m etim es e v en e x te n d in g t o a c e n tu r y o r tw o . T h is p r o c e d u r e , a lm ost u n iv ersa lly a d h ered t o in th e m a k in g o f h is to r ic a l d ic tio n a r ie s , is sh o w n o n fig u re 1."
            },
            "FIGREF1": {
                "type_str": "figure",
                "uris": null,
                "num": null,
                "text": "W e w a n t t o e m p h a size here th a t th e h is to rica l d ic tio n a r y , b e c a u s e o f its en o rm o u s s c o p e a n d sheer size, p o se s s o m e p r o b le m s w h ich c a n co n v e n ie n tly b e la b e lled ' o r d e r o f m a g n itu d e e ffe c ts .' T h e y a re n o t p r im a r ily p r o b le m s o f th e o re tica l d ifficu lty (a lth o u g h su ch p r o b le m s axe a ls o a b u n d a n t, as d iscu sse d in th e p a p e r b y J6n H ilm a r Jdn sson in th is v o lu m e ), b u t s im p ly p r o b le m s w h ich are related t o th e e n o r m o u s b u lk o f th ese b o o k s . T h e m a jo r aim o f th is p a p e r w ill b e t o d iscu ss s o m e w a y s in w h ich w e h ave a tte m p te d t o d e a l w ith th is p r o b le m . W e w o u ld a ls o like t o e m p h a size th a t w e are d e a lin g w ith a p r o je c t w h ich is in so m e re sp ects u n iq u e, sin ce th e w h o le e d ito r ia l p r o c e s s w ill b e c o m p u te r iz e d . T h is is in co n tra st t o p r o je c t s fo r c o m p u te r iz in g e x is tin g d ictio n a rie s , e .g . th e O x fo r d E n glish D ic tio n a r y (R a y m o n d a n d T o m p a 1 988) o r th e S ven ska A k a d e m ien s O r d b o k (R y d s t e d t 1 9 8 8 ). In 1983 a p ilo t e d itin g p r o je c t a lo n g tr a d itio n a l lin es w as u n d e rta k en . A re a so n a b ly fu ll d ic tio n a r y te x t w as m a d e u p , co n ta in in g w o r d s & o m a sm a ll su b se ctio n o f th e a lp h a b e t. B y c o m p a r in g th is sa m p le w ith a s ta n d a rd d ic t io n a r y o f th e Ic e la n d ic la n g u a g e (S ig fu s B lo n d a l 1 9 2 0 -1 9 2 4 ), it e m e rg e d th a t it w ou ld ta k e c lo s e t o tw o cen tu rie s t o finish th e e d itin g u sin g tra d itio n a l m e th o d s and p re sen t levels o f staffin g. A t th e tim e th is see m e d t o us an u n a c c e p ta b le w a y o f p r o c e e d in g . T h e r e w ere tw o rea son s fo r th is. O n th e o n e h a n d w e felt th e tim e sca le t o b e a b s o lu te ly u n a c c e p ta b le , o n th e o th e r th ere are w ell-k n ow n p rob lem s w ith th e tr a d itio n a l a p p r o a c h w h ich w e fe lt w e c o u ld d o s o m e th in g t o solv e b y fo llo w in g a n o th e r a p p r o a c h t o th e e d itin g task. 2 A Different Strategy B a s e d o n th e e x p e r ie n c e g a in e d in 1983, a n ew s tra te g y fo r th e ed itin g o f th e h is to rica l d ic t io n a r y h as g ra d u a lly b e e n ta k in g sh a p e in th e w ork ca rried o u t at th e In stitu te . T h is s tr a te g y sta n d s in ra th er sh a rp c o n tra s t t o earlier m eth od s a n d a p p ro a ch e s . It is c h a ra cte riz e d b y h e a v y re lia n ce o n c o m p u ta tio n a l to o ls. In d e e d , w e th in k it w o u ld b e fa ir t o sa y th a t w ith o u t th e ex iste n ce o f su ch to o ls th is s tr a te g y w o u ld n o t b e fea sib le. T h e c o n tr a s t b e tw e e n o u r n ew m e t h o d o lo g y a n d th e tra d itio n a l o n e can b e c a p tu r e d w ith tw o term s b o r r o w e d fr o m c o m p u te r scien ce . T h e tra d itio n al m e t h o d o lo g y c a n b e ch a ra cte riz e d as d e p t h -f i r s t , w h erea s ou rs c o m e s m uch clo s e r t o b e in g a b r e a d t h -f i r s t stra teg y. A d e p th -fir s t s tr a te g y m ea n s th a t e v e ry d e ta il in th e e d ito ria l p ro ce s s m ust b e r e s o lv e d t o c o m p le tio n as th e n eed arises. T h e r e is n o ch a n ce o f d ela y ed c o m m itm e n t, n o m a tte r h o w m u ch th e le x ic o g r a p h e r s o desires! S in ce th e ed itoria l p r o c e s s tu rn s o u t p a g e s e v"
            },
            "FIGREF2": {
                "type_str": "figure",
                "uris": null,
                "num": null,
                "text": "H o w m u ch m o r e efficien t it w o u ld b e i f it w ere p o s s ib le t o ca r r y o u t th e a n a ly sis in d e p e n d e n tly o f m a ld n g u p th e m a n u scrip t. T h is is a g o a l n o t readily a ch ie v e d , h o w e v e r, sin ce th e le x ic o g r a p h ic einalysis n eeds t o b e lin ked t o th e s o u rce s , i.e. th e c ita tio n s , in s o m e m a n n er. T ra d itio n a lly , th a t lin k has been a ch ie v e d b y in te rsp e rsin g th e a n a ly sis w ith q u o ta tio n s fr o m th e c o lle c tio n o f c ita tio n s .It is p r e cis e ly th is re lia n ce o n a lp h a b e tic a l o rd e r in g , co u p le d w ith th e en or m o u s w ea lth o f m a te ria l w h ich has t o b e a c c o u n te d fo r, w h ich giv es rise to th e a fo r e m e n tio n e d 'o r d e r o f m a g n itu d e ' effects. T h e fo llo w in g p o in ts are w orth m e n tio n in g :\u2022 T h e r e is a tim e la g o f d e ca d e s b e tw e e n th e e d itin g o f d ifferen t p a rts o f th e d ic t io n a r y w ith m a n y g e n e r a tio n s o f le x ico g ra p h e rs in v olv e d .\u2022 T h e e d it o r htis a v e r y lim ited v iew o f th e ta sk a t h a n d sin ce h e is la b o u rin g in a sm a ll c o r n e r o f th e d ictio n a ry . T h e a lp h a b e tic a l o r d e r in g o f th e te x t fo r c e s th e e d it o r t o d ea l w ith m aterial w h ich is in n o w a y rela ted . T h e s e p ro b le m s are a c u te b e ca u se o f th e size o f th e p r o je c t . W it h a sin g le v o lu m e d ictio n a ry , w h ich taJces p e rh a p s 10 y ea rs t o c o m p le te , it is re la tiv e ly e a sy to b y p a ss th ese p ro b le m s. N o t s o w ith th e h is to r ic a l d ictio n a ry . T h e research ca rried o u t a t th e U n iversity o f W a te r lo o o n th e c o m p u te r iz e d O x fo r d E n g lish D ic tio n a r y has b ro u g h t th is o u t (R a y m o n d a n d T o m p a 1 9 8 8 ). It tu rn s o u t, fo r in sta n ce , i f th e cross-re feren ces o f th e O E D a re e x a m in e d , th a t referen ces t o p re v io u s le tters o f th e a lp h a b e t a re m u ch m o r e fre q u e n t th a n th o s e t o su b seq u en t letters. M o s t cross-referen ce s a re, h o w e v e r, t o w o rd s s ta r tin g w ith th e sa m e le tter. W h ile th e la tte r is to b e e x p e c te d (m a n y cro ss-re fe re n c e s a re p r o b a b ly t o c lo s e ly related w o r d s ), th e fo rm e r result ca n o n ly b e e x p la in e d b y th e in d iv id u a l e d it o r ' s lim ited v ie w o f th e w h ole p r o je c t . A b re a d th -first stra teg y, o f th e s o r t n o w b e in g d e v e lo p e d a t th e In s titu te o f L e x ico g ra p h y , p ro ce e d s w ith th e e d ito r ia l w o rk fr o m th e t o p d o w n , b y m a k in g a n u m b e r o f passes th ro u g h th e c ita tio n c o lle c tio n , d e e p e n in g th e a n a ly sis a t ea ch sta g e. T h e r e are n u m erou s axlvantages t o su ch an a p p r o a c h : \u2022 S in ce th e e d itin g is c o m p u te r -b a s e d it ca n b e m a d e a va ila b le o n th e c o m p u te r a t an e a rly sta g e. \u2022 It is p o ssib le t o d ea l w ith co h e re n t p a rts o f th e v o c a b u la r y a t a n y o n e tim e. \u2022 It o p e n s th e w a y fo r d efin in g sig n ifica n t ph a ses in th e w o rk w h ich ca n b e finish ed in th e re la tiv e ly s h o rt tim e sp a n o f 5 -1 0 years. T h e first sta g e in th e e d ito ria l p ro ce s s , w h ich b e g a n in la te 1983, in v o lv e d a co m p le te p ass th ro u g h th e m ain c o lle c tio n , m a k in g u p a d a ta b a s e o f th e to t a l v o ca b u la ry . T h is d a ta b a s e is o f tw o fo ld use. In th e first p la c e it is o f tr e m e n d o u s va lu e fo r w o rk in lin gu istics, esp e cia lly th o s e areas d e a lin g w ith w o r d -fo r m a tio n a n d m o r p h o lo g y . T h e d a ta b a se has a lre a d y b e e n u sed t o in v e stig a te th e n a tu re o f c o m p o u n d in g in Ice la n d ic (K ris tin B ja r n a d d ttir 1 9 9 0 ). S u ch research is n o t o n ly o f th e o re tica l im p o r ta n c e , b u t is a lso relevan t in th e m a k in g o f p r a c tic a l la n g u a g e to o ls o n th e c o m p u te r , such as th e u b iq u ito u s s p ellin g -ch e ck e r. S p ellin g -ch eck ers are u su a lly d ic tio n a r y -b a s e d . D ic tio n a r y -b a s e d ch eck ers, h ow ev e r, h a v e s o m e lim ita tio n s in m o s t G e r m a n ic la n g u a g es w h e re th e p r o c e s s o f c o m p o u n d in g is q u ite a ctiv e. S o m e d ic tio n a r y -b a s e d ch eck ers n o w o ffe r c o m p o u n d w o r d p a xsin g (a n d o n e Ice la n d ic ch e ck e r is c o m p le te ly b a se d o n th e id e a o f w o r d p a r s in g ). D u e t o th e lim ited k n o w le d g e a b o u t c o m p o u n d in g , h o w e v er, it h as g en e ra lly n o t b e e n p o ssib le t o sta te th e co n stra in ts w h ich c o m p o u n d in g o b e y s , a n d th u s th e m e ch an ism s ten d t o ov erg e n e ra lize w ildly. S econ d ly , th e d a ta b a s e is o f g re a t v alu e fo r fu rth e r le x ic o g r a p h ic w o r k as it, a t o n c e , o p e n s u p m u ltip le search p a th s in to th e c o lle c tio n a n d frees th e e d it o r fro m th e ' ty r a n n y ' o f th e a lp h a b e t (s in c e a c o lle c tio n o f w ritte n d ic t io n a r y slips p e rm its o n ly o n e sea rch k e y !). N o w , h ow ev er, th e e d it o r ca n a cce ss th e c o lle c tio n o n th e b a sis o f g ra m m a tica l ca te g o r y , a g e o f cita tio n s , o ld e s t so u r c e , e tc. m e n tio n e d , th e first c o m p u ta tio n a l p r o je c t in v o lv e d a d a ta b a se c o v erin g th e w h o le v o c a b u la r y o f th e m a in c o lle c tio n o f th e In stitu te. T h is d a ta b a se h as 8 field s w h ich ca n b e b rie fly d e s c r ib e d as fo llo w s: T h e w o r d i t s e l f . T h e c h o ic e o f w o rd s w as p rim a rily b a se d o n th e o r th o g r a p h ica l fo rm . W o r d c l a s s . T h is , o f co u r s e , h as tr a d itio n a lly b een in d ica te d o n th e d ictio n a ry slips. A g e o f o l d e s t c i t a t i o n . T h e a g e is esta b lish e d t o th e n earest th ird o f a ce n tu ry o r t o a g r e a te r tim e p e r io d i f it is n o t p o s s ib le t o u n iq u ely p o s itio n th e c it a tio n in th is tim e fra m e ( o f 33 y e a r s ). See a ls o n o te b e lo w o n th e s o u r c e . A g e o f m o s t r e c e n t c i t a t i o n . T h is is c o d e d in th e sa m e m a n n er as th e a g e o f o l d e s t c i t a t i o n . N u m b e r o f c i t a t i o n s . T h is is n o te d e x a c tly fo r 1 t o 5 c ita tio n s . A ll w o rd s h a v in g m o r e th a n fiv e c it a tio n s are m a rk ed as su ch w ith n o flner d istin ctio n s b e in g m a d e . W o r d t y p e . T h is is th e o n ly ty p e o f in fo r m a tio n w h ich c a n n o t b e rea d d ire ctly fr o m th e c it a tio n slip s. W e fe lt, h o w ev er, th a t it w o u ld b e a d v a n ta g e ou s to a tt e m p t a r o u g h cla ss ifica tio n o f th e v o c a b u la r y a c c o r d in g t o w o r d -ty p e so w o r d s are m a rk ed as b e in g c o m p o u n d e d , a fllx ed , o r n o n c o m p o u n d e d . S o u r c e . F in a lly , th e s o u r c e fo r th e o ld e s t c it a tio n is n o te d . T h is is o fte n follow ed b y a n o te d e ta ilin g th e e x a c t a g e o f th e c it a tio n , e .g . a s p e cific year. In fa c t , th is field is b u ilt u p o f tw o su b field s: a b b r e v ia tio n fo r th e so u rce \u00a3ind re fe ren ce o f p a g e , e x a c t a g e, e tc. W o r d i n r e v e r s e . P a r t o f th e w o r d is kep t reversed in th is field. T h is is d o n e fo r th e in d e x in g o f w o r d en d in g s. A c tu a lly , w h ile th e V o c a b u la r y D a ta b a s e m ak es u p a re la tiv e ly sim p le d ic tio n a ry , it t o o k q u ite a w h ile t o p re p a re . T h is w as m a in ly d u e to th e en o rm o u s v o c a b u la r y in th e m a in c o lle c tio n . W h ile th e to t a l n u m b er o f c ita tio n s is a p p r o x im a t e ly 2 .6 m illio n , th e to t a l n u m b e r o f differen t w o rd s c o n ta in e d in th e v o c a b u la r y d a ta b a s e n o w sta n d s a t 6 0 8 ,2 0 5 . T h is sh ow s th a t th e n u m b er o f 'sin g le t o n s ' (w o r d s a tte s te d b y o n ly o n e c it a tio n ) is rela tiv e ly h igh , as n o te d b y J6n H ilm a r J d n sso n (th is v o lu m e ). W o r k o n th e d a ta b a s e s ta r te d in O c t o b e r 1983, w e rea ch e d th e 100,000 m ark in M a y th e fo llo w in g y ea r, a n d k e y b o a r d in g fin ish ed o n th e 7th o f M a rch 1986! T h e k e y b o a r d in g w as d o n e o n a V ic t o r 9 0 0 0 m ic r o c o m p u te r u sin g a sp e cia lly m a d e B A S I C p r o g r a m . A fte r th e k e y b o a r d in g w as c o m p le te d , p r o o fs w ere re a d , m o s tly in 1987. D u e to th e s im p le n a tu re o f th e d a ta b a s e , it w as p o s s ib le t o p e r fo r m n u m e ro u s in te g rity a n d c o n s is te n c y ch eck s w ith sp e c ia lly w ritte n p ro g ra m s. T h is c o n s id e r a b ly ea sed th e on e ro u s ta sk o f p r o o fr e a d in g . T h e d a ta b a s e re a ch e d its p resen t fo rm a t th e e n d o f 1988, o r five y ea rs a fte r w o r k o n it w as o r ig in a lly st^lrted. T h o u g h w e d id n o t k eep an e x a c t a c c o u n t o f th e w o rk in v o lv e d , w e w o u ld gu ess th a t it p r o b a b ly a m o u n te d t o a b o u t 1 0 -1 2 m a n -y e a rs. T h is g iv es s o m e in d ica tio n o f th e m a g n itu d e o f th e ta sk o f c o m p ilin g a tru e d ic t io n a r y o f th e m a te ria l c o n ta in e d in th e co lle c tio n s o f th e In stitu te ."
            },
            "FIGREF3": {
                "type_str": "figure",
                "uris": null,
                "num": null,
                "text": "A s a lre a d y m e n tio n e d , th e V o c a b u la r y D a ta b a s e c o n ta in s o v e r 6 0 0 ,0 0 0 re c o r d s a t th e p resen t tim e . P u t tin g th e d a ta b a s e o n lin e u n d e r th e M S -D O S o p e r a tin g sy s te m (w h ich has b e en o u r p la tfo r m fo r m o s t o f th is p e r io d ) w as n e v e r co n s id e re d a v ia b le o p t io n , sin ce w e felt th a t th is o p e r a tin g s y s te m w o u ld h a v e c o n s id e r a b le d ifficu lty in c o p in g w ith a d a ta b a s e o f th is size. T h is w as o n e o f th e rea son fo r a d e cisio n m a d e la te in 198 7 t o ch a n g e t o th e U n ix o p e r a tin g system .\u00d4 u r m a in co m p u te r s are tw o IB M 6 1 5 0 m a ch in es (p e r h a p s b e t t e r k n o w n as IB M R T / P C ) , ru n n in g th e A I X o p e r a tin g sy ste m . A t p rese n t th eir to t a l d isk c a p a c ity is 8 4 0 M b . W e a ls o h a v e tw o I B M P S / 2 m a ch in es ru n n in g A I X P S /2 w ith 230 M b o f d isk s p a ce . T h e R T s axe c o n n e c te d b y E th e r n e t a n d th e P S /2 s w ill a lso s h o rtly b e lin ked t o th e n etw ork . T h e N e tw o r k F ile S y s te m (N F S ) ru n s on t o p o f th e E th e r n e t c o n n e c tin g th e m a ch in es. T h e V o c a b u la r y D a ta b a s e ca n co n v e n ie n tly b e d e s c r ib e d b y th e re la tio n a l d a ta b a se m o d e l. It con sists o f o n e m a in ta b le , c o n ta in in g th e re c o r d s fo r ail th e w ord s, a lo n g w ith a s e c o n d a r y ta b le w h ich c o n ta in s in fo r m a tio n a b o u t th e sou rces u sed t o g a th er th e cita tio n s fo r th e m a in c o lle c tio n . T h e s e tw o ta b le s a re lin ked o n th e so u rce fields as sh ow n in fig u re 2. T h e r e are a n u m b e r o f re la tion a l d a ta b a s e m a n a g e rs a va ila b le fo r A I X , a m o n g th e b e tte r k n ow n a re O ra cle a n d In fo rm ix . D u e m a in ly t o co n s id e r a tio n s o f c o s t , w e d e c id e d t o use th e In fo r m ix d a ta b a s e sy ste m . T h is is a fu lly r e la tio n a l sy ste m w ith th e S Q L q u e ry la n g u a g e. U n fo rtu n a te ly , it tu rn e d o u t th a t I n fo r m ix has a n a sty p ecu lia rity^ in th a t it d o e s n o t a llo w field s w ith a v a ria b le n u m b e r o f ch a ra cters (VARCHAR). N o w th is, o b v io u s ly , m ak es life p r e tty d ifficu lt fo r th e le x ico g ra p h e r! F or th is rea son th e d a ta b a se g ro w s t o a p p r o x im a te ly 150 M b w h en it has been in d ex ed u n d er In fo rm ix^. W h ile th e d a ta b a s e is q u ite v o lu m in o u s , it is a lso q u ite fa st. It w ill in sta n tly fin d a n y w o r d a n d search on in d e x e d k ey s is a lso fast."
            },
            "FIGREF4": {
                "type_str": "figure",
                "uris": null,
                "num": null,
                "text": "ill n o t p a r tic u la r ly e la b o r a t e o n th e e x is tin g p o ssib ilities fo r sea rch in g th e d a ta b a s e . It is e v id e n t, fr o m th e d e s c r ip tio n g iv en s o far, th a t w e n o w have the p o s s ib ility o f sea rch in g th e c o lle c tio n in w a y s n o t p o s s ib le earlier. T h e follo w in g a re e x a m p le s o f q u eries w h ich h a v e b e e n p u t t o th e d a ta b a se b y m e m b ers o f sta ff a t th e In s titu te , as w ell as b y o th e r research ers. 1. F in d all w o r d s w h ich o c c u r fo r th e first tim e in th e w ork s o f th e 19th c e n tu r y p o e t J d n a s H a llg rfm sson . 2. F in d all a d v e r b s e n d in g in ' is ' ."
            },
            "FIGREF5": {
                "type_str": "figure",
                "uris": null,
                "num": null,
                "text": "3 . W h a t is th e p r o p o r t io n o f v e rb s in th e to t a l v o c a b u la r y ? 4. L ist all n o u n s fo r w h ich th ere a re m o re th an five cita tio n s in th e co lle ctio n , w ith e x a m p le s a tte s te d b o t h in th e 16th o r 17th ce n tu ries a n d in th e 20th cen tu ry . 5. L ist all n o n c o m p o u n d n o u n s fo r w h ich first e x a m p le s are a tte ste d in the first th ird o f th e 19 th cen tu ry."
            },
            "FIGREF6": {
                "type_str": "figure",
                "uris": null,
                "num": null,
                "text": "e have g iv en s o m e th o u g h t t o th e p o s s ib ility o f en h a n cin g th e V o c a b u la r y D a ta b a s e w ith o th e r k in d s o f in fo r m a tio n . A s d e s c r ib e d b y J 6 n H ilm a r J d n sson in his p a p e r (th is v o lu m e ), th e m a in th ru st o f th e e d ito r ia l p r o c e s s c o n c e r n s th e d e s crip tio n o f v e rb s. E v id en tly , q u ite a lo t o f m a teria l w h ich is o f re le v a n ce t o th e v erb s is filed u n d er o th e r w o r d classes in th e c o lle c tio n . T h is h o ld s, fo r e x a m p le , fo r p h ra ses a n d id io m s w h ich are o fte n filed u n d e r th e n o u n ra th er th a n th e v e rb . S o m e a tte m p ts w ere m a d e b y th o s e c o lle c tin g th e c ita tio n s t o file th em u n d er all th e relevan t ca te g o rie s, b u t it g o e s w ith o u t sa y in g th a t in a ta sk o f th is m a g n itu d e , s tre tch in g o v e r d e ca d e s, it is in e v ita b le th a t v a rio u s in co n siste n cie s o f p r a c tic e w ill a rise. W h ile th is e x te n s io n h as n o t b e e n im p le m e n te d , fig u re 3 giv es an e x a m p le o f h o w it c o u ld b e ca rrie d o u t."
            },
            "FIGREF7": {
                "type_str": "figure",
                "uris": null,
                "num": null,
                "text": "m a teria l w as b e in g g a th e re d fo r th e V o c a b u la r y D a ta b a s e , w o rk w as a lso b e in g ca rrie d o n in va rio u s o th e r areas re la tin g t o th e d ic t io n a r y p r o je c t as a w h ole. F o u r th in g s in p a rticu la r sta n d o u t: 1. T h e e d ito ria l p ro ce s s w as b e g u n in 1983 a n d g a th e re d m o m e n tu m d u rin g 1984 a n d 1985."
            },
            "FIGREF8": {
                "type_str": "figure",
                "uris": null,
                "num": null,
                "text": ". T h e In s titu te s ta r te d a c o lle c tio n a f m a ch in e -re a d a b le te x ts . T h e s e te x ts c a n re a d ily b e u sed t o a u g m e n t th e c o lle c tio n s . It sh o u ld ^dso b e n o te d th a t a lth o u g h m o s t o f th e te x ts are m o d e r n te x ts fro m p u b lish e rs a n d prin ters w e h a v e a ls o e n d e a v o u r e d t o a cq u ire o ld e r tex ts . S o m e o f th ese h ave been s p e c ia lly k e y b o a r d e d a t th e In stitu te .3. E x c e r p t in g fo r th e m a in c o lle c tio n con tin u es. A ll n ew c ita tio n s are n ow e n tered d ir e c tly t o th e c o m p u te r . 4 . W e h a v e a d o p t e d th e T^jX ty p e s e ttin g s y s te m fo r ty p e s e ttin g d iction a ries. T h is is fu rth e r d iscu sse d in th e p a p e r b y J \u00f6 rg e n F in d (th is v o lu m e ). F ig u r e 4 sh ow s in a s c h e m a tic w a y all th e m a jo r a ctiv itie s w h ere co m p u te rs h a v e b e e n b r o u g h t t o b e a r o n th e le x ic o g r a p h ic w o rk . T h is figu re sh ow s this a c tiv ity as it is c a rrie d o u t a t p resen t u n d e r M S -D O S a n d U n ix . W e are p resen tly in th e p r o c e s s o f m o v in g all th ese ta sk s t o U n ix . A n u m b e r o f p o in ts a re w o rth d iscu ssin g in so m e d eta il. T h e e d ito ria l p ro ce s s w as ca rrie d o u t u sin g th e R e v e la tio n d a ta b a s e s y ste m fo r M S -D O S -w h ich la ter b e c a m e A d v a n ced R e v e la tio n (C o s m o s 1 9 8 7 ). R evela r tio n is a d a ta b a se sy stem , b a se d o n th e P ic k O p e r a tin g s y s te m w h ich has e n jo y e d so m e su ccess in th e c o m m e r c ia l e n v iro n m e n t (R o c h k in d 1 9 8 5 ). T h e m a in rea son fo r o rig in a lly c h o o s in g R e v e la tio n w as th e a b se n ce o f a n y (s ig n ifica n t) c o n s tra in ts o n th e le n g th o f in d iv id u a l field s; u sin g , as it d o e s , c o m p le te ly v a ria b le le n g th fields. T h e m a in ch a ra cte ristics o f R e v e la tio n are su m m a riz e d in th e fo llo w in g : T h e b e n e fits o f A d v a n c e d R e v e la tio n are: \u2022 It is v e ry fiex ib le. It is ea sy t o re o rg a n ize d a ta files a n d r e c o n s tr u c t a p p li ca tio n s . \u2022 It h as a u ser-frien d ly in terfa ce. \u2022 It has va ria b le field len g th s. E a ch field ca n ra n g e fr o m 0 t o 65 k ilo b y te s , a n d p red efin ition o f le n g th is u n n ecessary. \u2022 It is p o s s ib le t o d efin e as m a n y as 6 5 k fields in ea ch r e c o r d , sind th e n u m b e r o f re c o r d s in o n e file is o n ly lim ite d b y d isk sp a ce. \u2022 It is p o s s ib le t o h ave m a n y files o p e n c o n cu rre n tly , a n d th ese ca n b e re la ted . \u2022 It h as its o w n p r o c e d u r a l p r o g r a m m in g la n g u a g e ( R /B A S I C ) , sim ila r to B A S I C , b u t m o re s tru ctu re d ."
            },
            "FIGREF9": {
                "type_str": "figure",
                "uris": null,
                "num": null,
                "text": "It is o n ly a v a ila b le o n co m p u te r s ru n n in g M S -D O S w h ich is a p rim itiv e o p e r a tin g system .\u2022 S in ce th e sy ste m has o n ly b e e n av a ila b le o n c o m p u te r s ru n n in g M S -D O S , d isk s to r a g e is lim ited . In sp ite o f th ese d isa d v a n ta g es w e h a v e fo u n d R e v e la tio n t o b e a sin g u la rly useful p r o d u c t fo r le x ic o g r a p h ic w o rk , a n d it is w ith s o m e sa d n ess th a t w e tak e leave o f it n ow th a t w e h ave m o v e d o v e r t o U n ix! T h e c o m b in a t io n o f v a ria b le len g th r e c o r d field s w ith a v e r y p o w e rfu l p r o g r a m m in g la n g u a g e h as tu rn e d o u t t o b e ideal. As of 1990 it is also available under OS/2. T h e N itty G ritty o f the Editing System W e w ill n o w d e s c r ib e th e e d itin g sy ste m as it w as im p lem en ted u n d er A d van ced R ev ela tio n . A s J 6 n H ilm a r J d n sson has a lre2 id y d e s crib e d in his p a p e r, th e e d itin g o f v e rb s p r o c e e d s m a in ly w ith re feren ce t o th e fo r m a l, s y n ta c tic 2ind m o rp h o lo g ica l b e h a v io u r . T h e e d itin g is b a s e d o n th e c ita tio n s a n d p r o c e e d s in a n u m b er o f ste p s. In e ffe ct, it is p o s s ib le t o v iew th e e d itin g as b e in g , t o a la rge e x te n t, a c o n tin u a tio n o f th e e x c e r p tio n in th a t th e c ita tio n s are p r o v id e d w ith g ra m m a t ica l m a rk ers. T h is is in c o m p le te c o n tr a s t w ith th e m e t h o d o lo g y tra d itio n a lly e m p lo y e d (K u h n 1 9 8 2 )."
            },
            "FIGREF10": {
                "type_str": "figure",
                "uris": null,
                "num": null,
                "text": "K u h n 's sorting board T ra d itio n a lly , th e e d itin g p r o c e e d s w ith th e e d it o r sp re a d in g all th e slips a tte s tin g th e o c c u r e n c e s o f a p a r tic u la r w o r d o u t o n a ta b le , a tte m p tin g to s o r t th e m m a n u a lly in to s e m a n tic c a te g o r ie s . K u h n has a n ice illu stra tion o f an 'e x c e e d in g ly u se fu l' s o r tin g b o a r d (sh o w n in fig u re 5 ) w h ich has b een u sed to a ssist in th e p re lim in a ry s o r tin g d o n e fo r th e M id d le E nglish D ic tio n a r y . It g o e s w ith o u t sa y in g th a t th e s tr a te g y e m p lo y e d a t o u r In stitu te is diar m e tr ic a lly o p p o s e d t o K u h n 's a p p r o a c h sin ce th e e d itin g p ro ce s s sta rts o u t b y a u g m e n tin g th e c it a tio n s w ith v a riou s en tries d e ta ilin g su ch fa c to r s as c o n ju g a t io n , s u b je c t , o b je c t , m e a n in g , e tc."
            },
            "FIGREF11": {
                "type_str": "figure",
                "uris": null,
                "num": null,
                "text": "T h e stru ctu re o f th e Lexicographic D a ta b a se as im p le m e n ted under A dvanced R evelation.T h e m e th o d u sed in th e e d itin g is a lso re fle cte d in th e s tru ctu re o f th e d a ta b a se p ro g ra m w h ich is u sed fo r th e e d itin g . F ig u r e 6 sh ow s in a sch e m a tic w a y h ow th e e d itin g takes p la ce . A ll c ita tio n s are k e y b o a r d e d b y a s e cre ta ry in to flies w h ich are th en im p o r te d in to th e d a ta b a s e sy ste m . O n e file keeps a lo g o f th e ed itin g p ro ce ss, n o tin g w h ich v erb s h a v e b e e n a n a ly se d , th e s ta te o f a n al ysis, a n d o th e r su ch in fo rm a tio n o f a g e n era l n a tu re . F u rth e rm o re , it co n ta in s field s w ith d e ta ile d in fo r m a tio n a s s o c ia te d w ith m o st o f th e cita tio n s. T h is file fu n c tio n s as a lo g o f th e e d itin g p ro ce ss."
            },
            "FIGREF12": {
                "type_str": "figure",
                "uris": null,
                "num": null,
                "text": "d th e e d ito r ia l p r o c e s s as su ch . A sp e cia l ' d ic t io n a r y ' con ta in s in fo r m a tio n a b o u t th e s tr u c tu r e o f th e files. E a ch tim e a w o r d is im p o r te d in to th e d a ta b a s e th e d ic t io n a r y is co n s u lte d reg a rd in g th e stru ctu re o f th e file w hich is t o b e m a d e fo r th e im p o r te d w o rd ."
            },
            "FIGREF13": {
                "type_str": "figure",
                "uris": null,
                "num": null,
                "text": "y d e sire d o r d e r a n d b ro w s e th ro u g h th e c o lle c tio n w h ile en terin gth e a n a ly sis, th u s d e a lin g w ith sim ila r c ita tio n s a t ea ch p o in t. A fte r ea ch pass th r o u g h th e c ita tio n s , it is p o s s ib le t o rea rra n g e th e cita tio n s u sin g th e fields w h ich h a v e b e e n e n tered . T h e c o m m a n d g iv en is th e s e l e c t co m m a n d .T h e fo l lo w in g c o m m a n d is, fo r in sta n ce , u sed t o o r d e r th e v e rb ' k o m a ' (c o m e ) on th e fie ld s fo r m , fo llo w e d b y v o ic e , fo llo w e d b y co n ju g a tio n : s e l e c t kom a b y fo r m b y mynd b y b e y g i n g T h is p o s s ib ility o f o r d e r in g th e c ita tio n s p r io r t o th e an alysis a n d a t ea ch su b seq u e n t s te p , is o n e o f th e m a in a ttr a c tio n s o f a d a ta b a s e sy ste m su ch as A d v a n ced R e v e la tio n as it en su res co n s is te n c y o f trea tm e n t a n d a ls o sp e e d s u p th e a n a ly sis co n s id e r a b ly . A n o t h e r a d v a n ta g e o f R e v e la tio n is th e fa c t th a t th e en try fo r m s h a v e b een d e fin e d in su ch a w a y th a t th e y c a r r y o v e r d efa u lt valu es fro m th e p r e c e d in g c it a tio n . W h e n th e c ita tio n s a re s o r te d su ch c a r r y -o v e r defa u lts a ls o serv e t o e a se th e d a t a e n tr y ta sk a n d en su re con sisten cy .F o r th e p e rs o n re s p o n s ib le fo r th e m a in ta in a n ce o f th e sy ste m it is v ery im p o r ta n t th a t ch a n g es t o th e s y s te m , e s p e c ia lly th e e n tr y fo rm s, ca n b e easily a ch ie v e d . A s p e c ia l fo r m s e d it o r (w h ic h , in cid en ta lly , g o e s b y th e n a m e ' p a in te r' ) m ak es it v e r y e a s y t o c h a n g e th e la y o u t o f fo rm s . It is a lso ea sy t o ch a n g e th e s tr u c tu r e o f th e files. W h e n th e a n a ly sis o f e a ch w o r d is fin ish ed th e sy ste m w ill o u tp u t a c o d e d m a n u scrip t. T h is is d o n e b y a sp e cia l p r o g r a m w ritten in R /B A S I C , R e v e la t io n 's p r o c e d u r a l la n g u a g e. T h is p r o g r a m uses a q u e ry c o m m a n d t o select th e c ita tio n s w h ich a re m a rk ed as s u ita b le fo r th e p rin ted d ictio n a ry . T o s u m m a rize b rie fly a t th is p o in t: \u2022 T h e e d ito r ia l p r o c e s s a u g m e n ts th e c ita tio n s b y a tta c h in g en tries to th e cita tio n s . \u2022 T h e s e en tries m a k e u p th e s o r t k ey w h ich is u sed t o d eliv er a u to m a tica lly th e m a jo r s tru ctu ra l lin es o f ea ch a rticle. \u2022 T h e d a ta b a s e 's \" re p o r t w rite r\" h as b e e n ch a n g e d s o th a t in stea d o f p r o d u c in g th e n o r m a l c o lu m n a r re p o r ts it tu rn s o u t m a n u scrip ts fo r w hich c a n th en b e d ir e c t ly ty p e s e t. T h e d iscern in g rea d er is n o w p r o b a b ly w o n d e r in g h ow w e g u a ra n tee th e in teg rity o f o u r d a ta b a s e , w h en o n th e o n e h a n d w e h a v e a c o lle c tio n o f cita tio n s , an d on th e o th e r h a n d cita tio n s c o n ta in e d w ith in th e e d ito r ia l d a ta b a s e . I f e rrors axe fo u n d in th e e d ito ria l d a ta b a s e , w h ere w ill th e y b e c o r r e c te d , in th e c it a tio n d a ta b a se , in th e ed itin g d a ta b a s e , o r in b o t h p la c e s ? O b v io u s ly , th is is a w eakn ess o f o u r a p p r o a c h as im p le m e n te d in th e A d v a n c e d R e v e la tio n d a ta b a s e sy ste m o f w h ich w e a re fu lly aw are. T h is , in fa c t , b rin g s us righ t u p t o th e p resen t d a te as reg a rd s th e d e v e lo p m e n t o f o u r le x ico g ra p h e rs \" w o r k b e n c h \" . It is o b v io u s th a t th e c o m p u te r iz e d c it a tio n c o lle c tio n n eed s t o b e in te g ra te d w ith th e e d ito ria l d a ta b a se . 5 Com putational Lexicography under U nix H o w s h o u ld th is d a ta b a s e b e im p le m e n te d u n d e r U n ix ? It is q u ite clea r th a t a d a ta b a s e sy stem like In fo r m ix is n o t at all u p t o th is ta sk sin ce it c a n n o t d ea l w ith fields h a v in g v a ria b le le n g th re co rd s . P r e s u m a b ly th e O ra cle D B M S w o u ld b e a b le t o c o p e axid th is is a lre a d y u sed fo r a t lea st o n e la rg e le x ic o g r a p h ic d a ta b a se , C E L E X in th e N eth erla n d s. H ow ev er, O r a c le is an e x p e n s iv e sy ste m , a n d , fu rth e rm o re , o th e r co n s id e r a tio n s h ave lea d us t o c o n s id e r a differen t a p p ro a ch . O u r e x p e rie n ce w ith T^jX, w h ich is a c o m p le te ly o p e n sy ste m w ith all s o u rce c o d e freely a v ailable, has b r o u g h t h o m e t o us th e im p o r ta n c e o f h a v in g p ro g ra m s w h ich are a v a ilable in s o u r c e c o d e fo r m . I f th e p r o g r a m d o e s n o t b e h a v e q u ite as desired o n e ca n alw ays ch a n g e th e p ro g ra m c o d e . T h is lesson has b e e n re in fo rce d w ith o u r e x p e rie n ce o f U n ix , w h ere an a b u n d a n c e o f e x ce lle n t so ftw a re in s o u r c e c o d e fo rm is a lso available. F or th is rea son , w e w ill b e m a k in g a serio u s a tte m p t t o cre a te a sy ste m w h ere w e ca n u se a va ila b le s o u r c e c o d e w h ich is (p r e fe r a b ly ) available in th e p u b lic d o m a in (e .g . a n d G N U ) o r c o m m e rcia lly . T h e d e ta ils o f th e a c tu a l im p lem en ta tion are cu rre n tly b e in g w ork ed o u t, s o h ere w e w ill lim it ou rselv e s t o an overa ll v iew o f th e to o ls e t, as w e h a v e cu rre n tly c o m e to v iew it. P erh a p s th e m a jo r c o n ce rn fo r us is th e fo llo w in g : S in ce th e c o m p u te r iz e d cita tio n c o lle c tio n is still b e in g a d d e d t o , w e n eed s o m e w a y o f k ee p in g a ccess to th e co lle c tio n o p e n , w h ile at th e sa m e tim e u sin g th e c ita tio n s , o r a t least a p a rt o f th e m , fo r th e e d ito r ia l w ork . N o w , b y re ferrin g t o th e d e s c r ip tio n g iv en earlier o f th e e d itin g p r o c e s s as it w as im p le m e n te d in th e R e v e la tio n d a ta b a s e , it ca n b e seen th a t q u ite a lo t o f th e in fo r m a tio n w h ich g e ts a d d e d t o th e c ita tio n s in th e ed ito ria l p r o c e s s is o n ly d e p e n d e n t o n th e c it a tio n itself. A s a m a tte r o f fcict, this h old s fo r 25 o u t o f th e 29 fea tu res cinalysed. It w o u ld , th e re fo re , seem n a tu ra l to d eta ch th ese fea tu res fr o m th e e d ito r ia l d a ta b a s e a n d s to r e th e m a lo n g w ith th e cita tio n s. T h is w o u ld lea ve th e c it a tio n c o lle c tio n in ta ct a n d still a llo w us to progress w ith th e e d ito ria l p ro ce ss. In th is w ay, th e fo r m a l a n a lysis w o u ld b e taken ca re o f.In this m an n er th e c ita tio n c o lle c tio n , th u s a u g m e n te d , c o u ld b e in p u t d i re ctly to a sort ro u tin e sim ilar t o th a t illu stra ted ea rlier fo r A d v a n c e d R e v e la tio n ."
            },
            "FIGREF14": {
                "type_str": "figure",
                "uris": null,
                "num": null,
                "text": "ideas o f how th e lexicographic database can be im plem ented under U n ix in th e futu re F ro m th is p o in t o n , w e n eed t o rely o n a sp e c ific a lly m a d e e d ito ria l p ro g ra m o r a c o lle c tio n o f p r o g r a m s fo r fu rth e r a n a ly sis. Such a c o lle c tio n o f p ro g ra m needs t o b e a b le t o h a n d le th e fo llo w in g : \u2022 It m u st keep a jo u r n a l o f th e e d ito r ia l p ro ce ss. \u2022 A s th e c it a tio n c o lle c tio n g ro w s, th ere m u st b e so m e m ech a n ism fo r n o ti fy in g th e e d it o r th a t fu rth e r c ita tio n s h ave b ee n a d d e d t o th e c o lle ctio n . T h e s e n eed t o b e in c o r p o r a te d in to th e e d ito ria l d a ta b a se. \u2022 T h e g e n e r a tio n o f m a n u scrip ts sh o u ld b e a u to m a te d . W o r k o n th e im p le m e n tio n o f su ch a p r o c e d u r e has ju s t b een sta rte d under"
            }
        }
    }
}