<&>Wellington Corpus of Spoken New Zealand English Version One <&>Copyright 1998 School of Linguistics & Applied Language Studies <&>Victoria University of Wellington <&>side one <&>6:26 i'm sure some of this is from him word afraid not okay um well all my updates are done apart from topic and word which i'm having to redo cos one of disks was screwed grant rang me up i mean kevin rang up today so that's just about done um i'm working on the bulletin um apparently i have a huge lot of cases somewhere um and building's fine as far as i know resource is fine there's not really a lot i can say about resources as i don't really know now word yet so i guess i'll find that out this week well um when you say you got a huge lot of cases this is it in <&>tape noise case bulletin no no just this backlog of old cases isn't it sal? yeah once we've immersed ourself in <,> and they're coming through production and <&>7:00 how much have you voc have you got voc much stuff to go into the bulletin um there's ten cases this week <{><[>and quite a few articles <[>so there's so there's ten new cases? yeah okay what's the backlog look like sal um it's in two lots the first lot er um a selection of mainly court of appeal and high court decisions which are only headnoted i discovered so they have to go in full text they have the first priority and then the second pile is just old cases just for a reference and they are not actually that important they can sort of be staggered out over the next few months <{><[>word word but the headnoted ones have to go in <[>okay okay and they are in production now they're in production now carrie's got them in separate files yeah and and there's about thirty all up oh in terms of the headnoted ones i don't think there'd be about thirty i mean it's a pile about that high <,> in terms of the content <,> some of it is fairly lengthy which is ah well i mean what we've got to do with <.>an <&>8:00 any sort of find out how much is in production <{><[>and and what the absolute priorities are because i mean jean's gonna tell us that her heap to go <[>mm mm <,,> and robert's got some as well <,> and there might be some cases i don't know word the case <,> you know see if it's at the health and safety's going on building this month but that's already through production so <,> um basically it'll be resource i'm focusing on have you got ledge <&>ie legislation coming through <{><[>brian <,> no? <[>coughs coughs <,,> all right um and and there's nothing sort of major happening with the act or anything like that there's only cases <{><[>where there r m a? <[>no no nothing happening there as far as i know the report of the planning tribunal's come out so that's going in we're gonna start putting those in on the database <,> that's all the only other thing is that during update this month i'll <&>9:00 be a little frantic because there's two conferences on between the sixth and the eighth so i'll probably start my update quite early <,,> these conferences are here in wellington yep building conference on the sixth and seventh and r m conference <&>noise the seventh and eighth okay <,,><&>4 um <,,><&>4 and bulletin's okay? resource bulletins? yeah oh i'm getting the hang of them <,,> now um have you guys had anymore <,> difficulties with um the winfax at all um well we haven't had one go out since last week's ones and that was so difficult the <{><[>one from the last job <[>the last word cos the <{1><[1>cables were <&>10:00 the wrong way round <{2><[2>wasn't it <[1>cables were the wrong way round <[2>yeah so um so i don't know as yet whether we've had any problems or not is that the problem that we had was only um a line it wasn't a part <{><[>being in right hole though was it <[>yeah yeah yeah once that was sorted out it was fine except what's that other fax is that an email thing um yeah i think so i'm not sure <,> exactly what it does <,> okay <,> um <,> jean well we're in the middle of photocopying out these <.>c tax cases to bring everything from ninety one up to ninety four in so that we got all the cases <,> what are we <.>ar like on ninety four yet do you know sal yeah i've done ninety four we start um eight missing from ninety four right so what does that bring our total to sixty? approximately sixty tax cases to go in to bring us so that we've got everything from ninety one up to and including ninety four on <&>11:00 form tax <,> from high court word still doesn't guarantee us b r s no word this is just going through <,> these and digging out the ones we're missing from here mm so we might need to go through the blue ones from matt's as well right you need to check that out as your next task sal just to make sure that we've got word well that's why you've got er <&>name to do is to um carry on with word indexes and completing the tax library yeah um and carry on writing up your current references your crossreferences too too far worse er policies the other ones they need probably won't have to do the full word word cases word that's that blue folder that jim showed you <{><[>sal <[>yep yeah i've got that word <{><[>and also to get copy word on butterworth's word <[>um money yeah money word <&>12:00 even better laughs word oh um so the the bulletin seventeen is butterworth's and brought out one that i think got's all the citations in it as well so <{><[>you'll want to look at that word <[>okay <,,> the copy you're working on how good is it for the photocopiers er i've blown it all up to a four it's pretty good did c c h scan from blownup photocopy they do <,> well yeah the photocopy quality looks pretty good okay um <,> well <,,> let's say tomorrow afternoon um <,,><&>4 just trying to figure out really there's no point just in sticking the scanner back out into production <&>13:00 unless we stick it on your machine cos the other the other one out there is grind to a halt with that new software you know it's really GOOD but er it needs a a class er four machine to run it still got the colour <.>one the other colour one won't run it? oh it'll run it it'll just be very very slow word <&>telephone conversation going on in background not transcribed really need to put that word with the new software with the new software <,> but <,,> so um sometime tomorrow afternoon um i wonder if we shouldn't get someone to sit down and play around by using that software on one of the machines so that IF my machine's available so that we can do it on word <,> or we can real pull the cord and plug it in somewhere else so that the scanner's available yeah oh that's a good idea <&>14:00 i take it you stick it in a four eight six well unless we um get a decent four eight six can you do it on a four eight six word my four eight six unless well it's an hour a day scanning at the end of the day i suppose yeah i mean there is that isn't there word well shall we do that no it'd be better on mine than on yours yeah <,> yeah i think we'll do that in the morning okay bring it across to there so then <.>wh who can try it out um me well carrie to start with but um should distribute those sorts of skills as widely as possible <,> so if i get carrie up to speed with that new stuff <,> the new software <,> tut um and then you can get production people up to speed with it and also um you know <&>15:00 once that's done er so distribute the skills through the editorial people as well <,> so IF the machine is available um and the editors sort of need something done <,> they can go on and get it done <,,> to put put it in its context these sixty cases represent what a couple of hundred pages <,,> yep how many have we got there any idea that's twenty <,> four yeah nineteen ninety three isn't so bad there were a couple of large ones in there and they were pretty small <,> i haven't started copying ninety two yet what's this what's the sort of scanning speed of it but ninety two going to be a similar sort of <{><[>pile to that one isn't it <[>um let's see word i don't know it's only a few seconds a page ten seconds um yeah for the to get the image up um you'd be <&>16:00 doing about <,> i mean <.>if when everything's optimised on my machine you'll probably be doing about a page a minute er carrie's which is faster you'd probably get that up higher but um the critical thing is er number one the print quality of what you're scanning and number <{><[>two you get the copy as close as possible to dead straight on the glass <,,> okay <,> anything else <,> tax four? <[>clears throat so so just to put that into context <.>w with that scanner running on a four eight six <,> we can move those reasonably quickly those cases yep so yeah you should be able to get basic good <,> most of the way clean text out of it at a rate of <{><[>about a page a minute <[>mm what's the feeling on which way should we approach this from ninety one forward? ninety <&>17:00 four BACKwards? <{><[>ninety four back <[>ninety four backwards right ninety four backwards um <,> so then the then the the big job from then is <.>j is proofing yeah but i mean again assuming that everything's um optimal the degree of proofing required out of it is er pretty um marginal because one of the things with that type of reader is that having scanned and o c red <&>ie checked using o c r software you can then go through and pick up on all of the characters it thought were questionable that it wasn't sure of <,> um and just get it up to <,> pretty close to a hundred percent quick spell check in word to catch anything that it THOUGHT it was sure of when it was actually wrong and then um straight into a final read <,> um <,> so it can be fairly quick <,> the only things that are gonna hold <&>18:00 it back are number one <,> the print quality of what it's scanning and number two <,> um how straight the material is on the glass <,,> anything else about the scanning or the tax four so can we be looking at sixty cases for this month um <,> tut <,,> no that's <{><[>all catching up as well it's not current stuff <[>just i appreciate <{><[>that <[>mm and it's not and it's tax only it's not anything of lana's i appreciate that okay i'd say that er yeah i mean we've just described scanning at at about a page a minute <,> and we're talking about two hundred pages there <,> so you know extrapolating that <{><[>yeah <[>yeah so that that gets us the clean text that gets you the clean text in a day er not oh cleanish text it still <{><[>needs proofing <[>in a day yeah <&>19:00 but er in a full day of doing nothing BUT scanning um you could probably <,> knock off well you could knock off a couple of hundred pages scanning then onto proofing yeah <.>i somebody could if the machine was available for a day oh yeah once it's done word <,,><&>6 see the other thing is that it can recognise on other machines then there's you know there's still time for reading on the other four machines to scan the images blow up the images and even the editors themselves could actually start something recognition without actually being connected to the scanner <,> so we're actually able to spread some of our word out yeah that's a <{><[>possibility <[>well this being what you were doing with <{><[>macintoshes <[>yeah <{><[>sniffs <[>mm i could do some saturdays and stuff because i imagine like <&>20:00 when my thesis mark <,> comes back through i'm gonna have to take some time off to <,> sort out some of that so i wouldn't mind building up some extra hours there okay well let's let's run a trial program then um with bulk scanning for just images <,> install the um you know stripped down version of the software on editorial machines and er see what sort of efficiency we can get by just get the images done and bring them out here but at the o c r but <{><[>for checking <[>i mean on that on that basis you know really over you know an hour a night for just three or four nights you'd get most of it in it's just bulk scanning and then if jean does it <,,><&>6 i mean we've got <.>to to push the production er <&>21:00 side as hard as possible because it's better to give it to absolute proofing er you know and they have better resources and that sort of thing than we do and that's exactly what word do scanning and stuff word whacking them in in huge quantities so you know we've got to set ourselves some pretty impressive targets in terms of the number of cases which you get in tax in terms of the number of cases and the number of t e rs we have and word <,,> okay we got any more to talk about with scanning and cases? if there's anything that needs to be picked up from varsity that i guess it's a case of putting that book back into operation sal so just scribble anything in the grad book and i'll do it when i'm up there <,,> okay <,> tut next sal <&>22:00 um <,,> nothing really the um oh a couple of these tax law went through third and final reading on the sixth of september <,> um that was the second reading of the bill that we put in the last update so we might get that through soon but there's no major changes in that um are third reading bills available to us that is have we checked that out at all no word <.>i just see how word that looks um exhales got a tax fax bills are going out this week so that's three cases for it three cases for it and then that's about it <,> oh b r s were we <,> <{><[>having that? <[>is our work continuing to supply the b r s <{><[>updates to law link for tax <,> we're not? <[>coughs word okay clears throat not when we word go <&>23:00 jean has just that topic <,,> er <{><[>so that's <[>and there's word needs to be done topic images should i have done the b r s update for building no thank goodness <{><[>laughs <[>so so what's the the push of your work then word seated in the chair i wouldn't actually say i had myself seated in the chair yet i spent last week doing the update yeah and i'll spend the next couple of days having a look to see what i actually have around the desk and then onto the other word um today i've just been working through the tax cases to get <{><[>jean jean <[>sure yeah yeah i mean and then it'll be the bulletin and then after i've got that out we'll go from there all right well perhaps you and i can talk at that stage yep once you get to there yeah and decide where we're going mm <,> you going word grant? <,,> i think it was just on and just on that you <&>24:00 know my priority <,> for for tax one is to get er as much case <,> and once it has references the case citation as possible is <.>that right yep so that you know as exactly the same as the as the er <,> er the pibs we were saying at least we could do a search of something if we haven't got it we take a word estimate out of the text yeah <.>a and that's the area we were missing is is all those cases <,> so i mean i think the first thing to do is to get all this case citations into tax one <,> sitting under the sections <{1><[1>and then having got that go back through and start <{2><[2>referencing as we start getting them in the full text so and you know it's gonna be quite a a big job to catch up to where jean needs to if she's gonna word <[1>word <[2>yeah oh shit yeah <{><[>laughs <[>yeah <{><[>word <[>and the other thing was <&>25:00 that er the way i look at the these databases word citations and everything else and mark them up on the different databases you're using at the moment gives access to to be able to sort all the word at the same time you might just pull out all the files really to see whether that might be appropriate for the <{><[>word <[>much like how we do our patch documents for <{><[>welfare <[>yeah for <.>m er grant and i managed to automate the process for that <{><[>word well i'm not saying all the pages but certainly just <.>h helping me through the word <[>mm no that's not quite clear <,,><&>3 that would be good um <,> tut anything else tax oneish <{><[>who's now doing tax two as well now or? <[>not from me yeah anything happening there? i haven't even got it on my machine yet so i don't know <{><[>if laughs <[>nah don't i had last month i went through and checked <&>26:00 out any g s t cases we were missing from c c h so we should be fully uptodate now how many about eight yeah okay they're all pretty weenie <{><[>yeah <[>that's that's good good <{><[>find <[>thanks for doing that lana no that's okay had to have something to put me up to um now what about um other development <.>e work in that area like um those er minor acts we were putting on like stamp cheque and you mean the cases for stamp and cheque no no <{1><[1>just all the all the ledge <&>ie legislation <{2><[2>word <[1>all the <[2>all the ledge <&>ie legislation is on except the only one we didn't end up putting on was the <{><[>a c c <[>a c c well that's fine that doesn't <{><[>matter <[>and that's the old a c c act the act the act itself so so we just sit on that so everything else is in yep <,,><&>4 anything else sal <,,> now carrie um i haven't got much to say really tax three's <&>27:00 struggling along <,,> it's just you know with that big push there aren't as many t e os and we're keeping up the chapters and tibs <,,><&>4 <.>s we'll sew that up is paula still working on all those t e os how's she going on them well she's not she just does the text word tidying up any chance of getting them word managed a temporary word <,> the tables are CRAP i mean it's just bloody useless <,,> so pushing them as hard as i can i looked at the laughs word we're got one of hugh's in there i wanted to show him probably before the <,> the next update <&>28:00 <,> and that's like a hundred pages of tables exhales tut it's table after table after table you know so i DOUBT very much whether we'll get those in word i'll have to skip whatever word to three in one huh but <,> so no i'm just plodding along really <,,> um <,> just so i'm counting a lot of flow charts and image chart things in that stuff oh word something <,> till he started taking word what do you want to do do you want to start catching up bringing those in no i'd rather get the rest of it out or you can just word <{><[>word word <[>yeah word the way those images are like word those ones? they're actually tax one <,> the discussion document <,,> <&>29:00 <{><[>okay <[>word is there anything we can do to speed up those t e os i mean <.>w maybe we should try um word there's noone out the back short of trying trying out you know the word advanced ware software that might be faster than the um wrap word scan i've got that on disk and it's still taking word we're gonna have trouble <,,><&>4 well now some of it's really word mm because really really word <,> and paula does all the tidying up for me word and i was just about tearing my hair out <&>30:00 well i mean is it going to be quicker if we start looking at those <,> images er before the tables or the flow charts and stuff? well both i was actually thinking no because by <{><[>the time word <[>i mean if carrie's spending all this time <{><[>sniffs cleaning up those can we use them word <[>yeah it might be easier for example to extrapolate all of the material that's in the tables all of the unique words the actual current word table instead of the actual table and then just taking in the image of the table word that image yeah i mean the way that we're doing tables now it's um <,> clumsy and a pain in the arse but um <,> with the things like you know if you've got tables that you change <,> um <,> er with er <&>31:00 relatively good size passage of text in the cell i e a couple of lines of text tut er <,> i don't know i still think it's worth having it all fully indexed in there so that phrase searches and so on will take you right there <,> and there's word stand a chance of getting you right there <,> but um <,,> yeah i'm going to have to revise our practices with tables somehow <,,><&>5 well keyword can move the software from the er you know six word for retracting the tables no cos it's still nothing that we've got here <,> er there's nothing that i know of in new zealand <{><[>that'll do tables properly <,,> no um even if there was er we're still saddled with <&>32:00 clients that are using stuff that doesn't do it <,,> <[>oh well you know the current rate of progress it's gonna be take us six months <,> to get these t e os in that are left and you know that's without any of the downtime as far as carrie's concerned <,,> the tables are all there in the t e os <,> you mean in the <.>fi in the stuff you've got on disk <,> um the files that you've got on disk er they're flat ascii or are they <&>32:40 <&>end of side one