Open Science and Collaborations in Digital Humanities Part 4

Video in TIB AV-Portal: Open Science and Collaborations in Digital Humanities Part 4

Formal Metadata

Open Science and Collaborations in Digital Humanities Part 4
Alternative Title
Data management
Title of Series
Part Number
Number of Parts
No Open Access License:
German copyright law applies. This film may be used for your own use but it may not be distributed via the internet or passed on to external parties.
Release Date
Production Year
Production Place
Dubrovnik, Croatia
Torus Group action Video projector Source code Workstation <Musikinstrument> Floppy disk Numbering scheme Open set Ext functor Shape (magazine) Disk read-and-write head Computer programming Mechanism design Labour Party (Malta) Different (Kate Ryan album) Negative number Videoconferencing Physical system God Theory of relativity Software developer Sampling (statistics) Shared memory Electronic mailing list 10 (number) Hand fan Data management Quicksort Annihilator (ring theory) Writing Reading (process) Directed graph Arc (geometry) Web page Computer file Civil engineering Variety (linguistics) HD DVD Student's t-test Computer Event horizon Frequency Differenz <Mathematik> Pi Term (mathematics) Operating system Data structure Metropolitan area network Computing platform Form (programming) Standard deviation Focus (optics) Wechselseitige Information Schmelze <Betrieb> Weight Expert system Line (geometry) Directory service System call Uniform boundedness principle Word Software Visualization (computer graphics) Universe (mathematics) Video game Game theory Family Freezing Library (computing) Building Information management State of matter Multiplication sign Direction (geometry) Decision theory Sheaf (mathematics) Mereology Replication (computing) Optical disc drive Mathematics Coefficient of determination Machine learning Hooking Ontology Data Encryption Standard Endliche Modelltheorie Thumbnail Scripting language Closed set Physicalism Knot Storage area network Self-organization Right angle Row (database) Server (computing) Implementation Link (knot theory) Virtual machine Web browser Repetition Wave packet Revision control Hardy space Natural number Internetworking Robotics Operator (mathematics) Addition Dialect Validity (statistics) Inheritance (object-oriented programming) Military base Cellular automaton Forcing (mathematics) Planning Icosahedron Database Peer-to-peer Subject indexing Voting Doubling the cube Compact Cassette Point cloud Enterprise resource planning Context awareness Dynamical system Execution unit Beer stein Water vapor Data dictionary Medical imaging Hypermedia Single-precision floating-point format Estimation Portable Object Adapter Error message Descriptive statistics Arm Content management system Digitizing Bit Demoscene Arithmetic mean Process (computing) Phase transition Order (biology) Mathematical singularity E-learning Spacetime Point (geometry) Sine Ripping Open source Patch (Unix) Diallyl disulfide Digital library Mass Code Product (business) Number Goodness of fit Intrusion detection system Representation (politics) Energy level Booting Computer architecture Scaling (geometry) Information Key (cryptography) Interface (computing) Bit error rate Physical law Content (media) Multilateration Cartesian coordinate system Integrated development environment Personal digital assistant Network topology Interior (topology) Code 40 (number) 1 (number) Set (mathematics) Insertion loss Function (mathematics) Web syndication Data transmission Usability Formal language Subset Web 2.0 Bit rate Synchronization Formal verification Diagram Covering space Area Algorithm Pattern recognition Electric generator Data storage device 19 (number) Entire function Application service provider Tablet computer Type theory Data mining Hard disk drive Website Resultant Thomas Bayes Cue sports Game controller Existence Backup Service (economics) Real number Information and communications technology Transport Layer Security Information systems Metadata 2 (number) Causality Tunis Fingerprint Noise (electronics) Dependent and independent variables Projective plane Interactive television Variance Search engine (computing) POKE File archiver Speech synthesis Object (grammar)
ok so owl what will going to do is compressed a law ste session him with the second hof of this session will the previous session if that make sense so are complete with the dead a menage meant and had ivor to jane the some mm a public ing erg ajman content yeah and ok so i'm some a as a said before some of this stuff.
might same common sensical at might be stuff that your he doer you but being explicit about it helps in your planning in thinking about menage ing a data so talk about the day to del you sz about small techne aches pursed structuring data.
how useful structures can be and then what to do with it preserving it and being sustainable with you data so a sick sz alee he www got more mo dated to work with him ill we already have the challenge of web ah cai's and the ther huge corpus of ste tex to work with but this is from twenty foot tainan it was scion that had dater is doubling in size every two year.
this and i'm than exponential kurth.
you've got more dated to work with all of the research partners an i mention previously you have to have a poke around or say if there's anything useful for you a whether it's easily these he to excess i'm coy or is a kind of aggregate research service that harvests all of the medard ed are from the institutional or pas a trees across the uk so this is traditionally.
ish unal re such out puts mostly but there's data at is starting to peer in nays and this is an aggregate search engine and if you look at was about three years ago or the has sixty six mill you know pp an excess articles the year ago one hundred thirty one million and their to hundred and that if i've million open access articles so or harry yi search through a as part of your or he search per i'm practice.
this is getting increasingly complicated and as a meant to do web ah ca i've is also something that challenging to work with where the working with the medard that are a lot owned all the text itself there is different requirements not just in of tyne in this stuff but computing it and what i guess a lot of your re such is for ck asst on his is an owl.
suh scap his how you've work with dater at that scale yet using mushing learning techne aches to in retune in hansen grav hold of the features that you're interested in i.
what whites important i think to you structures to do this is they can get very messy very quickly this is an example or of a re such project that didn't have any editorial control over a its for cab yul or he it just fill that metadata willie nearly it let research assistance use use the content management system used a rip ah za tree to fill it at new can see.
they got through entered foy six and fifty four tad agrees of spice and spy space server this is with a structured re such project that was creating mid addai out or not with a re such project that was mining annick struct ing medard out a from noise he text so it already the problem exists.
mechanisms up to control owes ugh really kind of editorial practices ole methodology zz for you so ste to employ which is restricting your vocabulary zz this is part of the quo the tiny process is you'll ger through in kind of a den to fight terms and words that mike cents and re form those are the time i'm but it's always was keeping a mind.
to constrain this is much as possible as on his it's usable for you.
and an example of how we use truck sher's amoebae rishi stran line which was the ri such projector talked about previously he is part of the ri blue a port project was its you are all structures was this one at the top which had soest that i ace picks question much pub i'd ache was seven three nine now seven three numb was just a dot a by side into fire it was designed by a person who thought about date of aces.
is in tens of databases they went really thinking about what that identify i meant his nerve relationship between seven three nine and the text itself ck so on murray built the saw it wheat deliberately an explicitly put a lot of eth it into using the structures of e u r o so become maini full structures not just for what the content of.
a the volya mm that the a den to fight his but with the structure within the roaring ms well so he using the anka tag an the your all to represent paragraph id ees to match through chirp eh are off i den to fi's in the hype dj team l. of the page so a if so one is now tsa do a digit a fly chai shun of this you are ol and the br issue stron line fight went offline.
they could stilt the rider from the text of the you aral what the actual printed source ch was that was used yet which you can't do when you just using new mehra khaled into fires so usable structures is something worth thinking about when your building models hw a traditional models but a very important and very usable i how many papal lls dess taw.
flook like this.
yeah a-k. yet say whoa a a and the br as a tad of thes yet yet so you like to par i'll your int for my shun and search through allayed up rather than filing your information and then finding a later yeah you nw so this kind of took roaches they were very use to the grew gul centric single k. would search the in time ass and make spec to get.
back what we expect to get back we don't really question whether or not the first ten results.
everything that we could find you know i with the filing in the medhat out a man ge hm unt and jane mention the importance of medard outer n't putting your research out online it makes it much mall rapidly discoverable i a up and this is a diagram from tim burn as lee when he was pushing for the link turpan data cloud and an oh you've had a session a.
and took and dater are ready but the id with this is that if you publish your out puts his paid e.a. f. it's a pry terry for my eye and its des fick look tick it into their out a can sourced tools that can extract texts from paid yoest but it's a lot more difficult to do then if you publish your data in see us faith ain the middle or in just plain text that's much more ck c'est of all.
for tools much mark cess uppal for people who know tools i'm much more ched or sul fa search engines to index end linked up and that are i'm on i think it's.
m m div alen about lng terp and ed or and and i think it's useful i think it's still slightly in excess of ol it's not quite the sayah aim as publishing opened tech stuart erp un data in a stand of forma this still have skills bury a for people to use link to up and dater and that's i think one of the challenges going forty's if you're publishing make to up and dale.
how do you make it usable but papal the daters self you put so now ce torre ihg preserve ation its astana billing this stuff is prob of the in a san from ilya how many paper hoodoo backup https of you'll work.
yes you nis d. is that a dropbox sink or is an actual back up.
i actual backup a-k. and you use you as be discs or cd rome the tes are another how dre ok good a-k. dues experience date a loss they've lost the hard disk i know jus i cakes hat ok.
right now it might sound silly if you haven't experienced it but when you do a your real as the importance of backup ce whether this sink back up sore additional ha jorrie ft backups a sin is you making of copy of something you having to maintain to the asians or wit so a a system in placed to mek an aai nz let his bakhat https.
ce where the going to cascade versions ove your research data direct tree that your collecting in building as part of you'll research is is up to you but there's mechanisms to help with that.
um how dries also a go out of star all yet this is a floppy disk in the universe iv london our car i have thw itch where and i able to get the dater off because we don't have a four than a quarter inge floppy disk any more and if we did have a farve earn a quarter int for feed ist can we put it into a modern operating system that could read the far system it was.
the put all of its medard outer on their it would in next little all so that it made it searchable in convenient there by destroying the actual bits on the discs a that is something about backup pts that is des fickle to the reais use in the future pp a at least you got them for disasters a eyes a whole other a host of alb salute media.
philly think these are ub slee as well i.
i know cds they a they a i know on a news it discs just a ca cess fg rekha a who years pts cassettes to store data that you did a s. they took a long time to get your date a off of them there the.
there it and and a that i'm not show how many cassettes with go and the aa car but again isn't obsolescence problem with the backups here and so it what it a the it.
yet way.
mmmm known and it's a des fickle problem so or yeah a yaffe tick a pp you all active research direct tree i'm accessible to you and backed out for you there's will its you you know i fs nut a it's a.
it does you in time and in storage.
so the talks of storage is well a lot of these your b. of e. see familiar with of the des jeux or or pause a dreesen ah car ised is whey ie you should really start to think about way you want to deposit your klain dup and finalise pts re such outputs i'm the uk web ah car if his sss well not a place to put your research day.
the and but perhaps the european open science clout is the yet and though be other clad him per structures that you'll tap into his part of you owe content discovery process is that you should think about reusing at the end of the all re such life orca liz own i m we've been exploring using archive matic a and preserve ih co which.
a digital preservation systems semde veh hr is another a version of those this an open source i guessed metadata rip ah za tree i'm the ones across the bottom soul owed is a a distributed storage system that timber in as les has been promoting recently and i p.a ther sees the into planetary fall system and dat.
these also a distributed date a perk a coal these are fairly new technologies then known wh widely used their distributor technologies so they're acta bit like a torrent to so you need to a laced have a host seeding your data for other papal to replicate and sing syndicate it a but i think these sorts of technologies will start to come into player than x..
five to ten use in terms of propagating and making researched outer available i and the reason i mention i'm is because the old technologies a going to got obsolete as they necessarily do so let's talk a little bit at documentation a using close for mites there's not much you can document in terms of accessing the content of the files and.
selves on this you got the applications tway pp a numb this is important so research of standards like t r i a in hay ch tino and hardy f. and the ontology zit you use so popular on toll a g.'s yeah an everybody can crate there are an on toll ajai but that's just crating another far law or so are a little bit of f. it in to figuring out which standards.
this or perp hr he a full oh the tools and the date a park ll un's that you producing goes along why documentation type server thes a century is part of the ri such project you've got a ministry iv methodological descriptive and technic cool book human tie shun but ministered of documentation goes back to the ri such project life sark as i was.
talking about an often though appeal like a gant cha pte where he just for would plan phases of your rhys a sheekey mm things it you much be looking at of a time there's dependency ce to these as well does anyone think about their research activities in this fe ease just whey.
as an e one planned head to their kind of second to furred year.
a little bit none it a ky so a mean you always have to be adjusting those you can never for plant too far ahead but it is useful to look as far head as you can given the information that you currently got and just be comfortable re forming the planned changing it completely or des.
it xing a completely his long as you're keeping document taishan for this then you can just a for your decisions of why you did what you did at that point in hte on of yuri search peruggia n't method a logical documentation this thi is mainly coming at in the form of jew put an oak books yet as your scripting if you just editing a text far all in your writing algorithms and scripts.
ce you're not necessarily keeping the versions as you got or but you quite your eye bl to do that much more easily in a jew put an opel ck his use ger pron a book's.
yen ucca their webbed browse a implementation of a programming environment so what you can do is makes cells of code with selves of texts with cells of graphics and chai pts and so essentially if you thinking about writing in a nut a physical nut book your processes explorations can do the same thing with.
pew taishan ol tench here.
i'm and these a a increasingly published on get hovers well it kind of shows people the thought process that show years how you were kind of in ter parading each of the algorithms as you went an it it's a trine of useful way to build things up.
descriptive documentation is more about kind of project level document taishan faial or date a by ce level book kummant taishan all variable or item of ol though kumon taishan and it's not quite the same is to the technical data its mall it to do wis the context of h. of these items we.
we thin you'll collection.
a i'm the uk data ah carve actually men dates some dated auc you in taishan when you deposit your data and if your thinking about other a pas a trees to put what your crating in collecting this is where these sort pox of documentation witcher additional files that have to crates ort additional work but it's wh well worth it and end.
some of the or pas a trees will mend i it that you include various to queue in taishan an now it's little bit like mehta date up its could be move a boasts you could just use a word dock human to kind of describe what you're thinking about your context it the time but there's a going to be really instrumental when you go back in one yea or three years' time to look at that directory or fadul zz that six hundred gigabytes.
weis and saw eyes and to inform you what you were doing with that at the time.
and the technical documentation a lot of this already comes on lawn and if you using open stand ids not boroughs then the technical stand ids in the sc aim a descriptions were lower a.d.b. of arla bei ol online but they would change over time because if you're using the libraries that have and i p.o. i for example than the i p ice technical documentation that the.
the ella bei ol online will be kept up to die it with the latest version of those live or ease and unit to make sure that those open source software provide as that your using.
hm will mine tyne aback cattle over of the versions of the a p eyes that you'll code is dependent on how mm if you're building something a self then you should put best f. its into describing technically the architecture is and the scheme is in the decisions fall you'll data structures.
i this is wes stand its helps that's not a stand of dusts just another tree relation dead abase a a i and who uses version control in their scripting or their da kumon taishan your you're ok do use it for over zhen ing your data the inn.
you use it for version ing your or project auc human taishan.
none ig can be used for all those things she et you can have private get rip ah za trees to keep snap shots ove your working research directories aback op is a snap short as well but you don't quite get this kind a replay ability with get for example you can kind of reverse a back to a previous state so if you completely.
for a cj should your over re such data collection whey you've had a one a fifty different p.d. f.'s that you've collected and groups nicely to get or and you restructured there's but you want to get back to that previous whey of describing things thin if you had kept all of your changes in version control than you could actually get back to that point to reen form use so ff of the.
these you might in the pos to.
thang courage you to learn get if you don't know itcz a heery aid ff.
so a it at a could to kind ud collection of data level it's really useful but at an individual document level is where really originate a from a was for source code management m four so ste would avella pp has to track in to in tame ce who with making changes to which lines of which far so are just by the night sher of this so that i could call woulda night.
they program of work and distributor the monks people in a tame what you've actually got is this really rich larne my lawn provenance model of who deve what to which lawn win.
so lousy to reflect on the pa ste work it allows you to river ch to the previous versions you know ott would we kapadia was has a version control if you were citing a specific page of which a paid year it a certain point in time then your setai shun will have when you access to it but that content of the page may have changed since then so.
least with we kapadia his website you'll the i would to or the person following out on your digital citation will be able to find the a precise text or www the page at the time that you accessed ould which could be important a m.
varying to troll ol as you to share with others you can synchronize copies between each other and the get preda call khan a manages a lot of that for you you can pour copies of a kurd base from someone else make your an changes to that and push there's a cross to them really easily you don't need to use you us be sticks quite so much so there is kind of a speed up with working and sharing collaboratively like this.
this i'm and as a much udell as utes is psych the price tate so there's a know though a into structure which is been book by cern already has version control so that you can cite specific varians all it not all rip aza trees will allow you to do this i'm that they should and and.
at and just a quick know on brie she stern lawn again there was a research i dicko but team hitch caught coal confronting the digital and what was happening in and jang correct me a fi character eyes issing couric plea is what was happening in historic oral research publishing was that research is we using digital rees or ce as like.
pretty she stron line to accessed texts and parm resources from the libre's in the aa cause it though interested in and they would then write their academic adi cools publish the but the citations that they we using hw are referring to the original pour kes to they could have conte to visit in the aa ca of some cell was but their way using the digital sarra get full and so are this.
so out ecl is really about the slight disingenuous nature of using a digital objects for your race a ihg but citing the original printed material itself so if you using digital objects if you using digital rees ol says couple at with version control in couple that with talk human taishan to in show all or that you cite the object that your literally.
the using not kind of using it oin ter printing it as a faraj get of the orb jeh qed when you des stott write a fining add des jeux a will rizo says as they prom resource as and sown.
and teach your so vz p it you haven't already.
yet this is a a link to the purred m. ing his store in web sought its of rhyl in os gentle introduction about how de ease get had a get the basic saw for up and running end than kind of methodology zz of committing changes to your code bases to your died abase as and then reverting in switching and back and forth.
the purred him historias quite a lot of other really great.
re such articles lots of technical ah tickle ff.
sutter on reproduce ability who all of this kind of data management one o one stuff is really to get you thinking about making what your collecting and what your creating as you do your research as a a kind of corpus that can be used for reproduce ability so the the more i open the more radically open about.
your methods your mistakes and yuri such dater in you docked mentation the betta.
this is where the kind of ck wanted i did methods come into it as well so this is the tro earth digital library which is used o.c.i. to scan and the a m a stray ll you news papers ih stark will stradey news pipers and you can see the icy aam eth as that they used are absolutely terrible yet they win proved finse but what they're built into this.
this was a crowd sourcing mechanisms to the people could comin incorrect it you but if you didn't nour the methods that they used to produce it an you would just using this is a search engine you wouldn't know what you went missing yet so there's no way to nerve what well you could look at the original printed material unsay which words these should bay pp but if you're a searching for the k. wor.
words and you didn't understand that they were errors in the ers she our production methods you wouldn't nour to dig a little bitty per for a the sorts of articles that you looking ful.
i'm in kuala tiny of methods as well kind of documenting these is very difficult are but this is an example or of a one of my calling says pay h.t. theses so yak sz lee had a parle of hype was like this which was all of the articles a-t. was using as part of his process and a that's not his falling cabinet buddy did girth through kind of filing process a man euro.
organizational process a physical pipers which is the coding process in call the tay www research yep that's much more rapid in much easy to do with tools like his are taro an engine urdd where you can just get a dio ought i for piper a nd druck put into use or tear id id by ace and taga and anna tight it kind of dynamic lee as you.
der i it and you it.
the it and be a a it that rees it your reich.
i that i a itcz yea likes that will the aunt that thanked the in like or he its let this it.
i he et man's men does ike and by much as actor butte don't know but mmmm be a it's a real you a that to your bracts will life it x. tick you know the writ think lel ff.
so i you owe.
it as it who's yat here i mean i think as jane mention one of the tools all ear is they inevitably get bought up by bui company and you lose the kind of openness and he thaw suv the mo's which is what might the poppy the des stott with i'm the terror is men tined its in depend.
and so far i hope it maintains that full longer yet it's quite good it'll thay will synchronize with this so is your entire i'm article collections yet in crean the files if you download the min attach than so it's it's quite a good way to kind a synchronize but again this is a synchronize i shun of a sub collection of your research using a separate two.
ooh sown.
i'm a the reason the we're talking about exposing the ghats of old of yuri sich is for critical inquiry really allows other people to interrogate you'll articles at the end of ol of ease and his an example from a call ache who did you used a a used pall i'm entry medard out a language to run sentiment an owl a fifa.
crossed the speech texts from the uk hants odd corpus now a exactly have our and which software he used his written in his research ah tekel but the implementation of the software hal would implements centamin a now asus i'm which lexicons were dictionaries it uses to do that is something that you need to then ff.
make a further inquiry to figure out i'm thes then on top of that he's kind of tack yul a tidd and agra grated the sentiment across the speech is a cost the years and this is nowhere in his arctic cool that gives an idea of where the source code for this much bay and ove the oud arra them as that he's used.
to to do this to to ce tooele kak you leyshon that's produced this visualisation soeur what i'm getting at here is if you want to to follow ought with this and interrogate the methods of an article you an a to contact the researcher os plym questions i ask for they dar us see if you could get as sample of their algorithms all made.
the them to talk through the methods in person if the not explicit in the published output so and his ove isay the pens on the kind away that the ri such articles are written as well so this is miss a de logical but it doesn't going to technical methods think goes into a kind of hire level ska matic meth a.
so i think about that is your writing your articles as will people will come inquiring for you and can kind of get them away few more the boson explicit about have you've done things the way you've done things and they the its he caught jt the he it a let's it and it.
yester he and i are and.
and soul of this is really about the scientific method it's about the reproduce ability of the ri such that you're producing you know i'm but i guess is a distinction between replicating an experiment and reproducing the findings of the experiment if you completely publish you dater in he completely publish asst source code this still uno woman of getting that up an runny.
in for somebody there is a as soft were development in ck eaker as system the is you knit you can't just pick up any all librarian stott running it on york computer there is orb v.c. that pine the have to go through to get scripts running undo use open sore soft way so that's not quite replication of the experiment that that's a hurtle that most people will have to go through.
i'm reproducing the findings would be figuring out exactly haj our our description of what them methods hw are getting the original source data preps getting thes the same libre's at they were using and then reconstructing that for you self to try to replay the experiment rather than kind of re running the entire a source kurd a mend.
critter bove us as transparent is ste khan of a a gets back to the critical inquiry he is so this article is scribble we could in kwai out as much as we want tatoo to figure out from the ri such as a published this how they did it why they did it and what their findings work along the way i'm but if they had published all of that.
on the bloke who on a that or oppose a tree in different forms then that's kind of mo to woods transparency so that if the all of those objects find ub ol undiscovered will then we kind of heading to what's transparency rather than require in papal to be scrutinising everything in a manual way so uf facilitates the ri.
if ft of your research.
i'm and the confronting the digital hm i tickle here is the one that i describe previously about citing the digital object rather than they regional force material and gen is mention digital object i dent afar as as will so as many year of the eye tums that you publish if you.
ca tache a dude you object identifier to them then at least to kind of improving the find ability in discover bill ity of your hr easels of.
a-k. crown sourcing a m let torre quickly about the us zz a f. ect iv lee its outsourcing of small ta six it's the the pace work for the annotation work it the values labour there's a lot of negative connotations associated with it but of its used properly and your trance peyron about how you've used to and.
more you've used to then it can be are really useful tool i and ove risi use it for machine learning and a taishan.
i'm so galaxies oot is a crowd sourcing war rizo ce i launched in two thousand and seven as ooh nw averse a well they've got also projects nme here can talk bet a the ina www or ce some of the project's ons in averse yet a and they sewing ck.
a sheekey and mill you he may a had into mill yer is in of s. in the galaxies he projects it was little inch which is that the very c'est crowd sourcing which ste to identify and ass in awe mikel fe ge cross this what kaing diff i nag bia or sell mm was being represented in the pages and i signed up for it when if eh started to thi how it wirtz and it was.
this fund tass tickly boring off to you'd lick to the bout five of these pitchers there was nothing kind of ing gauging yoon there are soul but they they didn't need me to stain do a two hundred is say got you know a million people doing one he ch and then going away them that was falling but that was the first i.d.a. on they've maeve to mutt schain mall what into you and humanise ease reset sheekey and the cre a.
they did jenner ayesha nw of text fall at for people like us to do research with mnd the handwritten tux recognition it's improving all the tar line that the couple of years ago even it was it was really reaney paul so the best way to jenner right ok create content was to get people to chime scribe it for you and so the images of the documents of be.
up online and then they would ask i am people to vall an tay air at to lan ind transcribe and caret ext fw and dade try i'm hugh late between peoples contributions so you didn't you to be annex pair at if three people transcribed the same letter the same way them they would take that as a result but if gif there was differ ince they'd wait intel another pasts nut come along to tram sc right the it.
irv mm nd and it was very much pitched as crowd sore saying is are you aim wh wonderful academic cover great idea ear don't have an of toying aim a you people out there can do if uel me which is where this kind of x. ploy says to ft asp at to that most he riz talking about comes they mm ter it and i think some of the project so you're going tsk are all mm to talk about here all much mall and this idea all of share.
airing or price s. as i'm coia cre age teen dater and working with the people he might be doing their pssst to help as generate khan times you an pts tree.
and so one of those praa ger ix is called low is of london project it's a public a heritage from dud project in it pts focus was on public can gauge munt an there was a creative agency the designed a this were up the cation but mood the design of the a pluck ation was explicitly about allowing school children to contribute content and and a take.
cent and behind the scenes they have an edge tory a process making scheuer that what has been contribute is still relevant and quality controlled.
uh huh is us just an of the scree chalk of that i'm a mentioned the tro oeuvre digitize ation of the a the strel you news papers i'm again they've got welp it was a very reliable initially but they a ft kind of bill tin these kratz sourcing mechanisms to a lier aaj people to hannah tight it what i'm not sure about is they.
ting processes for those submissions so anyone could go in there and and a tight and correct the o.c.r. but you have to dig around in they website to see how they approve those changes in with or not they verified by human and a mm some other examples with that kind of got are around crowd sourcing were a the early english books on.
in line ybor t.c. pay which used double re cane process which ease ascension he talking it twice doing a difference between there's to text far lls and checking for any difference as where there's differences look back at the original pie ge can ce to mike the correction that supper priya and these methods an auntie not we sen on tune on point nine percent accurate.
it the much more costly much more labour intensive i'm but you kind of get the ecj or see out of it other transcribe bent thump rassas it the university college london launched in two thousand and ten and explicitly tried to build in a game a fide mechanism for transcription wh with a verified approval praagh ce.
cess for what people will contributing and so they've got the bent thumb oh mutter i'm since twenty tend they've trance groped twenty two thousand pages and twenty one three a twenty one thousand of those nine e six percent have been checked in approved by the it a tory ol tame soeur part of the crowd sourcing is not just using am as on.
work to a get the labels that you want they zz of verification a validation process by people her know the content you can veh or four and vouch for the the labels that your eye bl to pif ful on mass.
i the row society is recently launch they web saw it the science in the making which is also in attempt to do these collab urdd he eve krauts soest.
veh hr far i'd content publishing systems a are this uses a ce a series of peer of you mechanisms behind the sainz in it ministry to www user into faces to make sure that what people look contributing conform not just it a tory a god line so they mocked up correctly but they're actually not just spamming the system an hoping that it gets published on the side as rep he pte a sz.
and management involve not just in academic call it he control.
a and i was auma ten ih go kercher oven mechanical took is probably use fulp but again it's the diminishing the value of labour our ends these kind of a difficulty in hell you credit or attribute the papal who are helping you i'm and if their incentive is soley financial they going to be kind of trying to ratchet as many label.
o.'s as they can wih that much care for the content that their labeling and if you need to use mechanical turk for.
the a mount of and a tay shuns that you nays you a least nate document that this is al you got them.
up there's a nother a tool pp a which is not creds sourcing pp per se but it's a hand running transcription tool arm which has used mushy learning technologies from paley org refers hua handwriting experts to inform the annotations that we used for the.
a machine learning modal zz the a used for hander writing or kg mission so a this is a case where you do need experts to do your annotations and you could crowd sore some if you want to to put i think pill yer gruff as a kind of difficult to cross source.
and and the sorts of examples that come up with this is the british me's rishel arbor he released a whole bunch of its image as on to twitter and the grey tags at the top were i think the years yeah s. so human generated tags wh once this collection of images was eve.
really ste on to twitter the human generated tags with n. used with an image classifier to generate the blue tags which is the share locked net tax and someone as les would this is a will the based and a good new ub but the machine is come back an call that a dog or so it using fail it accurate.
human tags with stew what errors in the mush ain classification ta sc so it's des fickle had a scrutinise the reason that these of kamau n't i think the best but to do that is to be open about the models and the annotation tools that you've used so the people could attempt to reproduce a wh is another example it's a koala it's been tagged what a medic.
three as a doke meant does have a dog form butts.
someone is misspelt koala k. you ale a i sir these are oh a and this is a simple verification process that your you want get with.
a hwat i'm as on turret for example i think you might need to per scribe quite specific guidelines to the sorts of tags at they use but again you need to put effort into making show that you annotations a correct not you that it it it it it you you it you i that that.
now you gaal a it and he and i you that is he a that really to the it will they but a and thawing did nair it is it knows the thick going a i'm thirty are not thought jun a he will our way rye did za guy wrong pts a ites it's the what dick nery a gers they earn or g.
it let it in now more fact he you with x. people guy stuff o j me and the importance of the pier were v process.
i owe i'm a think that ste it for crowd sourcing so a its it's a useful tech naik but you need to bill the in you project management meth ids shore peer revue processes and your veh hr for cation.
so on thing i and of the tuna ge nw.
and the.
he ha an we would guy nw to get easy deer in ak sys now we had a ellis to various and krantz though zing platforms i'm we were going to get you to have a go in see what its loy eke to be on the news this side of pfi and how the different interfaces affect the quality if the information the ear going to get from people but i think given that we it where i aver running off to le.
and chewy a.a. we worry sss do that will provide the list of sight say you can go an have a look out the mnd at there are various ones for trans karting text for marking arp architectural plans to identify the shape of the building that kind of thing there lots of different forms and but we we white do that now will was skip aver that and gaze straight on to the next.
session which is a little bit about public in erg ajman dh and then we going to hmm nd buck t. you for the last ex ce size of the day with and elevator pitch which i would explain a little bit later jb fw we've taught a lot today an unit yoon have heard surrey talking with he was soup of ices that one of the really important aspects of any of these year pym porsche.
jeh qt is to showy the white of isley fusses sys see and to communicate the importance of your wort qc a to very specialist public's but also to ud general pub lake and people hugh know nothing about your work atta tool gk so the national cool gia nineteen stand to for public iin erg ajman to in the uk described that kind of i t of itty.
the as the miri i'd have wh ways in which the activity and benefits of higher education and research can be shed with the public img ajman is by definition a two way process to that comes bought some of the crab so thing things the marty ris to king about involving interaction and listening with a go lael of generate sing mutual ben a ff.
sets thay when it works well it's not just about you telling some i'm what you do on then thinking i've done my public ing ajman and going your way i ideally you'll learn from their responses to your work as well i'm they will have learnt something from you that they can take away and it will help them to think different me about something so it's not transmission of.
may shun it's the exchange of knowledge and different knowledge from different public's and it's very hot to get rights i've been in knots the of public you gauge when event slum show marty us the fly aim were and i could m. it comes the lois fans of the front gets what's affective lee an arc a time it lecture and then leaves off to t. questions butts not public engagement i'm.
arts knots and a fact if way ft earring it however good des speaker ith to ha earn and i think it's him put a n't to thing could bout the different audience be taught a righty about high who you are you trying to gets to a because what good of public ing erg ajman pte looks like it's going to varian norma slate depending on the groups of people that you'll talk.
heme to so a they ste is a a and a chart again from that from the same body in the uk wish describes the different kind of audience is and stake cold as for their or search that you'll doing soeur they have public sector it waun ends civil society the general public business and very subsets of that in cuil eating social and to prize.
this cultural in ce to chea you shun gns community t a i like commune a c. and nongovernmental organizations depending on the area where king in i would say that probably at least two o three if these gurry https particularly fit for the project here business and creative in des stree would be.
first id in the kind if work the yurt earring and you need to think differently about the best way to get them to listen to what you want to tell them.
earn you may own know about the european research is nights i think this shows how well im bet it this a's in these e.u. funding prairie grams at theirs is an ewell e.u. research is night which is designed to show you cased the types of research the dipping funded by the year a pained commissioned and its to i don't know if any of you've got plans to be.
in vote they ce year it's come quite early in your search but it's some the twenty eight to september so they may well be something going on in your eye own he in ce to chea shunned i am certs the good own good eye days to go alone in and see what other people a doing and what was fee you and what you might be able to learn from how they ff presented it sheekey and again if someone sting something you thing all that's really.
great just cope ear it it's could practice nabi mines he.
or in my own in ce chea shun the yoon of s. t. www london we run and am ewell national festival if the hue manatees called being human which in vall owes institutions across the country this no millie about three or four hundred different institutions taking part and we give it us thien every year it's so that people you know this something.
in to it to hook p pull and get them in trusted and this year its discoveries and secrets than that's running in nev behm ber and we've done a number as ike tiff a tease aver the years to public yn erg ajman arounds digital research mostly using web auc i've zz in the very first year if the project which how to des jeux.
tittle thame to air pts at what it means to be human in it to tittle aei a ch we presented the web ah car i and it was pretty nw you've that and i and i was pretty new to nine what web ah kice were all most nobody he came alone to the event have the thirst idea what web ol ck carse were am we had a loader discussions along the lines of why also you all cars the hole where.
that welt lime not the whole web but you know quite a base of that and this is heart wort kes and this is heiss to front from the life wire bei he and it was really and i just the nature of their ees kinds of prey grams we had ques a lot of retired people came along he ryn not bat familia even with the loi i eve way i bei it a but it was amazing how quick in they got in hte rested when you could.
showy them for exam polled this is their website of your village sz cer sz ten years ago a and there is something in it for than that they could recognize and get a honeyed lawn and could him usually see the importance and value as their sss and got it grating and to standing of why it was important reserve digital material ce for the future and if.
for a thing we found it easier to explain to that group of people have an to i q dem ih kes but oh i'm what the value of this station tool or zte real a lot his store in psych own now i.v. on that kony ease that are where ers a memes of the public which yell this is great out of can i see the company that i you stir uk for ten years ago so it's fine t. not i nme gold that it's going to get people oat.
cit remember what you're to king about us.
and we also in a in a following year in upon a shit with our light per he ran a computed a video a gain all kaif as n pts using the at the competing games that has been our kite and then than emulated from the internet archive and a and that we pssst we the sud so the six o seven.
an nde games that we setter at running it was a bit if a challenge for us because they had to be emulated on the internet all ck life to ste not tell you what them you key start ix that replace the old operations are so you had to kind of dick trial an era and go he sir ever thing and say it ha o. k. a control l. means that that carried to punch asst somebody or mmmm i so a how.
the bit difficult especially if you've never plate the gains be fall which against beaks but to that documentation and all say digital eth no grow fate if you'd got someone in he was a real este expert him using what of thi exe they could have said immediate late a-k. yr that carrot to space titty they section and so were but all of that's me sing erg said we had to reconstruct it i so he got people aired nw.
to replay these games heart a really boy nd variety of audience is a most just cheetah abroad all of his am maass to students a long they were doing competed design on they were very in hte rested in the eats at a old a us that ick i we had kwai suh lot of school till ge earn coming him with that parents and the parents could rehman s. about playing me's games when they we young.
abound annette was muta the children as well and again imparting night dear about the value as these it's important items that cultural heritage and andrs and it we also tried to make it look a little bit like kier am nd and auc eight is much as three could it's so or even how to couple of disco bowls gehring to got the lighting around ict and reeks such.
just online floor that what i would cool material coach of but the posts to a xers and a mag a-z. eames and so on from the thai nw that talked about these and we printed order those up put numb around said that people could see the context for them as swell and the the most popular game with one of the oldest won.
and switch is an american educational guy which she kunz the over an the bush of mare could the oregon trail on it was designed is the kind of when if they scarfed way games where you make you decision and that such your ft down one written then another won to see house excess or you would a pain of navigating across to the west of americ care if you were poire near i think i.
made it to the and juan pts but my in toy a family drown dull mm the way a pert you know hey the sings happened or mmmm if but you know you might a have an a food or you i didn't have an of bullett so if theo with i kind of thing and we had a greek oaf pp maass to students he who spend a good to you our is with this read he rida mench or he game wh working three to get to the and.
and i getting it under ston the your for internet culture was like then and how you katou u z xm a tyrus pet education so they kite it can what really whalin it's great fun us well having people air nw and getting really excited about the kind of stuff that you work ck with anna in tress to tui ths ck at for this year's events or nouri this is going to be connected up.
to clear patch or a set will have clear patch or bran day ing at we're doing something court our mutual friend the machine arm each ooh friend being a dickens norful am nd a we running it pts n when not making people come to a euna var thes tain at people who you haven't gone to yoon of us t.a. it's a md universe is can be.
or in tim is a ting setting foot oh you've rudd bought dual when you think it's nata place fee you is a real challenge soay were a we want to take things outs to people i mean it it's not exactly aei yary some mother you can to get probably lots of middle claw ste people coming any wake as its to the berkshire hope he in the middle of lumb de and but it's still in a temp to mewed and.
you could whole things in libraries oh community send to psych kind fang so we can a ho to and the bought in the basement of wh water steins buik schulp in top and court rate in cent chilled london and matti spin exploring and competed generated poet sher he from there and nineteenth century cope its from projects arm get can who hine this elves get in burke pte sorry gang rim.
an route and nd a year we know i loud vis a to ce to can cib yeas seed words and see what poetry they get and back from that sss and were oath they can have an actual para it tooth going to come alone am going to have what we call a perch we sly i am which the contest between ang nd the machine jenner ak shud poetry and the actual poa it and see which people prefer a on then get in to think about.
orth a shit what does it mean when a perry mns been produced boy machine learning whose the all for the person who out him he created the algorithm the algorithm what you d. about the dater it was trine don't so what does all ther ship me nw zz that ats lacan of aka dem ih ck aspect that we want people to experiment with a amazing.
and and this isse mn of the pair him use the that marty's generated but the your the authored or the elgar them's the ought then eve sid that from the pyre mns of robot browning and it it's so wata of was kes a say was aai a nd shut social be the rest he won the ce crowded bench its knell shull lead the throbbing train or take the memon pts fr he.
freeze the wayda nd the cloud the clad ailed know so i my cho arch unal imei chill arch we a to you claisse lee links today tick law nz with moat ist klein this on the crowd but point you wa kes i many put it yeah i'm there but it's id it that a you neri little bit klien key i wouldn't if repeated the clough or to the cuil eh ailed to myself but they you get a a moon.
we founded it what's much much better was poach or eight at the praise you can immediately tell that it's not been purty ste by hemon because the graham astrup justa to can pick a should knit bright stam really quickly ck and it's not logical where his and perch she doesn't need to be logical a in fact it's its obscure or on i'm big juris by ne ge.
schist say we cui we can has thumb from with bart eye and and really get people thinking about machine there he well do the joe you sats exactly what we going to do yeah absolutely were going cert to give beast times at to people an are some who to you think crate their sti thing hit was a pair it would he think it with his artificial intelligence he boy.
so go meant an that you know grun stir loot it rooper you'll there'd room a gal that yet you yap and he had and and one of my khani some you're talking about its immediately said well if ale i brights poetry's it interested in breeding et i own you think all i kang he hit that's a complicated for the so quick cookware ce jun rich re probably.
we die it once ago a what did i i want to to reid's para trait or we sit a force feeding the algorithms where they stay to fit can rights it does it want to rida ould a he thet owais you know i the you did a interest income as say shuns that you might not expect twenty eady this con of wire gk said a were and say you can i think mate really quite challenge ain and.
re search ick sauce uppal if you've read he think about earned have caught a little bit of resource behind it slow and and it's really could from to do it's well they bearing in mind that idea of making your search excess apoe to people owe what we'd light you to spend at the rest us this the shunned daring ace t. come up with an elevator.
the pitch for you'll research i don't nif you've heard about elevator pitch is that's quite american business soar to thing that it's descried at the short to description as an idea it product or company that explains the con set in an way that any lists nw a can understand in a short to period of time this description typically explains his.
here you the thing use fall what it does why it's need or to and how it will get done.
fw say read light you to spend ten o fifteen minutes thinking about your own wrist search and if you was standing in from to have an exam a group of sick sting to eighteen year olds how would you disc roy eve your research why it's important and why they should be in tresses in it.
the in and stan present it to everybody sauri about this in in won to two min is mo were lough loan elevator rides of rick couple of minute its seo have a think about the way that you would explain et to it gri that not just of non specialists but oh of people in that the last stay to second dri education how which you.
your podge it to cross to them and particularly why it should matter to them white should be interesting sheekey so with no he specialised langridge ch its theri a care and gif you kind of ten to fifteen minutes and then ascii to come up and percent but to ever puttick a yet you nour yeah i buy.
you'd fs sss are carried eyfs you i.


  336 ms - page object


AV-Portal 3.19.2 (70adb5fbc8bbcafb435210ef7d62ffee973cf172)