Author Archives: admin

Dealing with Disks

Posted on by
Reply

One very specific aspect of digital Preservation is ensuring that the contents of physical disks can be accessed in the future. I found I had four types of challenges in this area: 1) old 5.25 and 3.5 disks that I no longer have the equipment to read; 2) a CD with a protected video on it that couldn’t be copied; 3) two CDs with protected data on them that couldn’t be copied; and 4) about 120 CDs and DVDs containing backups taken over a 20 year period. My experiences with each of these challenges are described below:

1)  Old 5.25 and 3.5 disks: I looked around the net for services that read old disks and I eventually decided to go with LuxSoft after making a quick phone call to reassure myself that this was a bona fide operation and the price would be acceptable. I duly followed the instructions on the website to number and wrap each disk, before dispatching a package of 17 disks in all (14 x 5.25, 2 x 3.5, 1 x CD). Within a week I’d received a zip file by email of the contents of those disks that had been read and an invoice for what I consider to be a very reasonable £51.50.  The two 3.5 disks and 1 CD presented no problems and I was provided with the contents. The 5.25 disks included eight which had been produced on Apple II computers in the mid 1980s and these LuxSoft had been unable to read. I was advised that there are services around that can deal with such disks but that they are very expensive; and that perhaps my best bet would be to ask the people at Bletchley Park (of Enigma fame) who apparently maintain lot of old machines and might be willing to help. However, since these disks were not part of my PAWDOC collection and I didn’t believe there was anything particularly special on them, I decided to do nothing further with them and consigned them to the loft with a note attached saying they could be used for displays etc. or destroyed. Of the six 5.25 disks that were read, most of the material was either in formats which could be read by Notepad or Excel, or in a format that LuxSoft had been able to convert to MS Word, and this was sufficient for me to establish that there was nothing of great import on them. However, one of 5.25 disks (dating from 1990), contained a ReadMe file explaining that the other three files were self-extracting zip files – one to run a communication package called TEAMterm; one to run a TEAMterm tutorial; and one to produce the TEAMterm manual. Since this particular disk was part of the PAWDOC collection (none of the other 5.25 disks were), I asked LuxSoft to do further work to actually run the self-extracting zips and to provide me with whatever contents and screen shots that could be obtained. I was duly provided with about 30 files which included the manual in Word format and several screen shots giving an idea of what the programme was like when it was running. LuxSoft charged a further £25 for this additional piece of work, and I was very pleased with the help I’d been given and the amount I’d been charged.

2) CD with Protected Video files: This CD contained files in VOB format and had been produced for me from the original VHS tape back in 2010. The inbuilt protection prevented me from copying them onto my laptop and converting them to an MP4 file. After searching the net, I found a company called Digital Converters based in the outbuildings of Newby Hall in North Yorkshire which charged a flat rate of £10.99 + postage to convert a VHS tape and to provide the resulting MP4 file in the cloud ready to be downloaded. It worked like a dream: I created the order online, paid the money, sent the tape off, and a few days later I downloaded my mp4 file.

3) CDs with protected data: I’d been advised that one way to preserve the contents of disks is to create an image of them – a sector-by-sector copy of the source medium stored in a single file in ISO image file format. This seemed to be the best way to preserve these two application installation disks which had resisted all my attempts to copy and zip their contents. After reading reviews on the net, I decided to use the AnyBurn software which is free and which is portable (i.e. it doesn’t need to be installed on your machine – you just double click it when you want to use it). This proved extremely easy to use and it duly produced image files of the two CDs in question in the space of a few minutes.

4) Backup CDs and DVDs: The files on these disks were all accessible, so I had a choice of either creating zip files or creating ISO image files. I chose to create zips for two reasons: first, I wanted to minimise the size of the resulting file and I believe that the ISO format is uncompressed; and, second, on some of the disks I only needed to preserve part of the contents and I wasn’t sure if that can be done when creating a disk image.

Having been through each of these 4 exercises, there are some general conclusions that can be drawn:

  • The way to preserve disks is to copy their contents onto other types of computer storage.
  • The file size capacities of old disk formats are much smaller than the capacities of contemporary computer storage formats. For example, none of the 5.25 disks contained files totalling more than 2 Mb; the CDs contain up to about 700 Mb; and even the DVDs contain no more than 4.7 Gb. In an era where 1Tb hard disks are commonplace, these file sizes aren’t a problem.
  • There are three stages in preserving disk contents; first, just getting the contents from the disk onto other storage technology; second, being able to read the files; and third, should the contents include executables, being able to actually run the programs.
  • The decision about whether you want to achieve stages 2 or 3 will depend on whether you think the contents and what they will be used for, merit the extra effort and cost involved. In the case of the 5.25 disk containing TEAMterm software described above, providing a capability to run the application would have involved finding an emulator to run on my current platform and getting the programme to work on it. I judged that to be not worth the effort for the purpose that the disk’s contents were being preserved for (to be a record of the artefacts received by an individual working through that stage of the development of computer technology).

Listening to New Stuff with Alexa

Posted on by
Reply

Back in February, I reported on my attempts to get Alexa to play the albums in our music collection. I’d found the following:

Coverage: about 80% of our albums were present in the Amazon Music Unlimited library.

Specifying Discs and tracks: for albums consisting of more than one disc, there appears to be no way of specifying that Alexa should start playing Disc 2 as opposed to Disc 1; and, similarly, there’s no way of getting Alexa to play a particular track number.

Voice Recognition: Alexa couldn’t recognise about 10% of the Artist/Title combinations even though I had checked that they were actually available in Amazon’s Music Unlimited library.

Since then I’ve been using Alexa and Amazon Music Unlimited to listen to newly issued albums reviewed in the Guardian/Observer newspapers, and now have a further substantial set of experience to compare with my original findings. The first thing to say is that being able to listen to complete albums, as opposed to just samples of each track from Amazon on my laptop (as I have been doing previously), is, obviously, a far more rewarding experience; and to be able to listen to a range of new releases from start to finish, regardless of whether or not they suit one’s innate preferences, is a real luxury. Most I will never listen to again – and some I have cut short because I really didn’t like them; but there are a few which I’ve really liked and have made a note of at the back of our ‘Sounds for Alexa’ book. At least I now feel a bit more in touch with what sort of music is being produced these days.

Now, to get back to the topics I covered in my earlier findings; below are my further observations on each of the points:

Coverage: Since last February I’ve checked out eleven lots of review sections comprising write-ups of 121 albums. Fourteen of these albums were issued in CD format only, and all the other 107 albums were available in Amazon in MP3 format. All but nine of these 107 were advertised as being available for streaming or available to ‘Listen with your Echo’ (the latter being the Alexa device); and of these nine, six did actually play through the Echo device.  Of the three that didn’t, two would play only samples (Bob Dylan’s ‘Triplicate’, and The Unthanks’ ‘The songs and poems of Molly Drake’); and for the other one (Vecchi Requiem by Graindelavoix/Schmetzer) Alexa repeated “Vecchi Requiem” perfectly but said she was unable to find any album by that name. Given that only three items were actually unavailable, I conclude that a lot of the new albums that are being issued in digital format are available in the Amazon Music Unlimited service.

Specifying Discs and tracks: It still appears to be the case that it’s not possible to specify that Alexa play the 2nd disk in a two disk album, nor to play a particular track number. To get round the multiple disks problem, a number of people in the Reddit noticeboard suggest creating a playlist in which the two discs are listed separately. As for the track number, Alexa will step through the tracks if you keep saying ‘next track’; but, if you really do want a particular track played, the best way to achieve that is to use the name of the track when requesting it – both of the following worked for me:  ‘Play Kashmir by led Zeppelin’ and ‘Play Cromwell by Darren Hayman’.

Voice Recognition: Of the 121 albums I checked out, Amazon claimed that 98 of them were available to play through the Echo, whereas, in fact, I could only get 85 of them to play. For eleven of the other thirteen albums, Alexa just couldn’t understand what I was requesting; and in the remaining two cases, Alexa a) insisted on playing “Rock with the Hot 8 Brass Band” instead of “On the spot” by the Hot 8 Brass band, and b) played Mozart‘s Gran Partita by the London Philharmonic instead of by the London Symphony Orchestra. Turning to the 85 albums that did play through the Echo, it was significant that only 59 of them played at the first time of asking. For the other 26, I had to repeat the request at least twice and as many as six times (these details are included in this Recognition Analysis spreadsheet). Naturally I was trying out all sorts of combinations of all or part of the particular album title and artist. After much trial and error I have taken to first asking for both the album title and the artist (play me X by Y); then, if that doesn’t work, to ask for the album title on its own (or even just parts of the album title – for example, 1729 for the album title “Carnevale 1729”). Finally, as a last resort, to just ask for the Artist. This strategy proved successful in all but 3 of the 26 instances that didn’t play at the first time of asking. These figures indicate that Alexa’s voice recognition capabilities haven’t improved much since my last write-up in February. This view is reinforced by my (undocumented) experiences of trying to get Alexa to tell me about various golf, rugby and cricket events. Her responses have usually been either about a completely different event or just that she doesn’t know. Perhaps I’m not asking the questions in the right way….. at least Alexa is usually able to provide a weather forecast at the first time of asking. In her defence, I should mention that my son seems to have no trouble in adding all sorts of outlandish things to our Alexa shopping bag (which, I should add, we don’t use – Alexa just provides it if you want to put things into it).

From this summary of my recent experiences with Alexa, it seems that little has changed. Whilst Alexa’s voice recognition capabilities don’t seem to have improved much, the usefulness of the device compared with having stacks of CDs around, is undiminished. So much so, in fact, that we have replaced our last remaining CD player, which was in the conservatory, with  another Echo device; and we’ve upgraded to Amazon Music Unlimited for 10 devices at £9.99 a month.

There are undoubtedly many other uses that we could be putting Alexa to – the weekly email from Amazon always suggests several new things that one can ask her or get her to do. We haven’t really followed any of them up. Perhaps I’ll get to printing out the email each week and putting it next to the echo as a prompt. Or maybe I won’t  – we’ll see.  One thing’s for sure: what with all our CDs in the loft, and no stand-alone CD player, Alexa is going to be with us for the indefinite future.

Throwbacks

Posted on by
Reply

The inspiration for this thread for Ideas came from a paper-based Ideas Book which I set up in 1972. It didn’t really get many entries and some of them were more reflection than specific ideas; and it’s lain dormant for many years. So, I’ve just scanned and destroyed the physical Ideas Book; however, for completeness, I’ve recorded below some of the items from it (suitably summarised where necessary).

Throwback 1 – 08Jan1972 – The idea of an ideas book

I guess the first idea to go into this Ideas Book must be the idea of having an Ideas Book. Basically, I think that, although thought is of paramount importance, thought without action is a great waste, both of time and – yes – ideas! So, in future, if, sorry – when, I get some crazy idea – something completely original and far out (like establishing a World Tune Library with a cataloguing system based on every possible combination of notes over, say, a two minute period; when a new tune is sent in to the library it is played into an analogue/digital computer, and this would then produce a ‘catalogue number’ – it would be most interesting to see just how many more tunes were available at any one time), then I will write it down and it will be on record to act upon, elaborate, or even just to read over and laugh! Something quite amusing about this Ideas Book is that maybe the only idea I ever put into it that ever gets acted on will be this first idea to have an Ideas Book….

Throwback 2 – 11Jan1972 – A light design

A light-come-ceiling decoration system could be constructed out of hollow cylinders made of stiff white paper of varying lengths – width 32cm with 2cm of that used for overlap and the varying height resulting in holes of about 10cm width at various different levels. A reflector could be made by covering a sheet of stiff card with bacon foil (to which the vertical cylinders could be secured) which could be fixed to the ceiling using aerofix)

Throwback 3 – 20May1972 – The soundproofing stubber

The vast quantities of cigarette stubs that are wasted could be used as sound proofing material by manufacturing attractive boxes which have a stubber through which the tip would be released into an inner, cheap and recyclable, container which could be removed and sent to a sound proofing company. Profits could be made from the sale of decorative external boxes and from the sale of sound proofing material made from used tips. Stubbing would be cleaner and more efficient; and there would be a reduction in cigarette tip pollution.

Throwback 4 – 21May1972 – Investigating the warping point

Our experience of the world tells us that there is a causal factor for everything. When I look at the stars I think about who or what put them there, because logic informs me that there must be an answer. As I think about this question I assimilate all the relevant information I have until there is maximum capacity thought but an inability to provide an answer. The result is a split second of total confusion. It would be interesting to see what electroencephalograph readings appear when this point – the warping point – is reached. I wonder if other people have the same experience, and, if so, would the measures vary depending on the level of comprehension that different  people have of the question? Would the measures change over time as people increase their comprehension of the question?

Throwback 5 – 01Jan1984 – Simultaneous phoning and computing

It would be useful to have a unit that would interface between a Type 96A jack plug and a home’s telephones/computers. The unit would enable, at the very least, the simultaneous use of the telephone and the use of the computer over the networks.

Throwback 6 – 01Jan1984 – Game designer CBT – and the potential for progs

A computer-based training program could help children design the logic of a computer game i.e. the design specification prior to programming. A program like that could sell for £1 a time. With the right kit at home and a link to the networks, you could design, build and test such a CBT program in the space of 24 hours at home and be selling it immediately over the networks. If it was a novel and good enough idea the mass network market would soon provide 25,000 purchasers; so you could have made £25,000 within 48 hours of first having had the idea.

Throwback 7 – 28Dec1996 – Crucial pursuits

There are five crucial pursuits for members of the human race:

  • Making other individual humans feel good through love, tenderness, intimacy, caring, understanding, and good deeds.
  • Creating the conditions for other humans to have better lives.
  • Learning and understanding about the world and universe about us and about our fellow humans and the way we live.
  • Learning and understanding the origins of humankind and the important findings, discoveries, secrets and developments that humankind has made and encountered.
  • Learning and understanding the origins, secrets, and meaning of the universe and its relevance to ourselves and humankind.

The Verdict

Posted on by
Reply

Back in April, I asked 6 friends to pass rapid judgement on my latest attempt to define the Roundsheet application.  I asked them to give the document a quick scan and to provide answers to the questions below with either Yes or No, and to feel free to add comments or suggestions if they wished.

  1. Were you able to understand what was being described?
  2. Do the Roundsheet concepts make sense to you?
  3. Do you think a Roundsheet application would give users something they haven’t got already?
  4. Is it worth pursuing this idea any further?

I received 5 replies. Not everyone was able to answer either Yes or No to each question – some answers were ‘partly’ or ‘sort of’ or ‘don’t know’ or ‘possibly’; so, I’m going to classify all such in-between kind of responses as ‘not sure’. Applying this rule of thumb, the bare number results were:

Question Yes Not Sure No
Able to understand? 3 1 1
Concepts make sense? 3 1 1
Something user’s haven’t got? 2 3 0
Worth pursuing? 2 2 1

 Some of the comments were interesting:

“one would think that many spreadsheets have pie chart functionalities: your concept is really about how to use those functionalities”

“whether this could be a protected tool? I think you would have difficulty”

“still wondering why the roundsheet format is any better than the tabular format which apparently could be used instead?”

“may be of interest to those who get frustrated by the complexities of linking Microsoft Excel and PowerPoint applications – I have come across a few of those in my career!”

“I feel there is a core there that could be extremely useful”

However, overall, there’s no overwhelming consensus that this is a winning idea, and it’s probably unlikely that time spent trying to promote its development into a product would be rewarded; so I think this is the time to put this journey to bed. I have enjoyed the intellectual challenge it has given me; and have the satisfaction of knowing that I took the ideas as far as I could and finished the job. I could always bring the topic out of retirement should someone come along with a serious interest in taking it forward.

Thank you to all of you who, over the years, have taken the time to wade through and pass comment on the various specification documents.

NORNOMOT (nora-no-mo)

Posted on by
Reply

Sometimes you hear about people who are always invited to events but never host any themselves. Similarly, some people don’t respond to communications or Xmas cards; and it’s not uncommon for presents sent to growing child relatives to remain unacknowledged or thanked for. In all such situations the giver begins to feel a little aggrieved with the situation, but perhaps feels it inappropriate to raise the matter directly with the individual concerned. To assist all those in such circumstances, it might help if there was an unobtrusive but clear way of signifying dissatisfaction.  Perhaps a code could be attached to the bottom of an address or invitation in the same vein as SWALK (sealed with a loving kiss). I suggest NORNOMOT (pronounced nora-no-mo) standing for No Reciprocity (or Response) No More Of This. Maybe the greeting card manufacturers could create special NORNOMOT cards which include pictures of a Last  Chance Saloon.

Hikes through the preservation hinterland

Posted on by
Reply

I’ve just finished dealing with two particular digital preservation challenges that exist within the document collection I’m currently working on. The first involved two Lotus Notes files; and the second concerned some Windows Help files. My experience with these issues illustrates a) how just a few files can take a lot of work to resolve, and b) that there’s often an answer out there to seemingly impossible preservation problems provide you are prepared to look diligently enough.

I really didn’t believe I was going to find a way to unlock the Lotus Notes files since Notes is a major and very expensive piece of software that I don’t possess; and, in any case, it applies sophisticated time-limited password and encryption controls for its use. Despite being aware of these issues, I thought I’d take a quick look on the net to see if I could find any relevant advice. It was time well spent; I discovered that it’s possible to download a local evaluation copy of Notes for 90 days, and that, because it doesn’t run on a server, this sometimes enables old Lotus Notes files to be opened. I duly downloaded the software and installed it; and then, regardless of the mysteries of Notes access controls, had access to the whole of one of the files (which contained conference-type material) and to parts of the other (which contained sent messages). I still had the username and expired password from the time the files were created and I think this may have helped to access the latter – though I’m not sure about that. Anyway, in both cases, I was able to print out the material to PDF files. I had to manually reorder the conference-type material and to reinstate a few hundred links in it, but that was it – job done!

The Windows Help files were a lot more demanding. Microsoft stopped supporting the WinHelp system (.HLP files) in 2006 in favour of its replacement, Compiled HTML Help (.CHM files). Although Microsoft did issue a WinHelp viewer for Windows 7 in 2009, WinHelp is essentially an obsolete format – it isn’t supported in Windows 10. I’m still running a Windows 7 system so am still able to view the HLP files – but they had to be converted now if they are ever to be accessed again in the future.

There is much material on the net about how to convert HLP files into CHM files, but, as someone with no knowledge at all about how files in either of these systems are constructed, I didn’t find it easy to understand. I soon realised that converting from one to the other was going to be a challenge. However, I did eventually find a web site which offered clear practical advice which I could follow (http://www.help-info.de/en/Help_Info_WinHelp/hw_converting.htm), and I duly downloaded the recommended HLP decompiler; and the Microsoft HTML Help Workshop software. The process to be followed went something like this:

  • Decompile the HLP file into its component parts (consisting of a help project file with the extension .hpj, along with one or more .rtf documents, an optional .cnt contents file, and any image files – .bmp, .wmf, or .shg – that are used within the Help file).
  • Convert the various HLP files into HTML Help files using a wizard in the HTML Help Workshop tool (the new files consist of a project file with the extension .hhp, one or more HTML files, a .hhc contents file, an optional .hhk index file, and any image files that are used within the Help file).
  • Set parameters in the hhp file to specify a standard Window name and size; and to have a search capability created when the files are compiled into a single CHM file.
  • Reconstruct the Table of Contents using the original HLP file as a guide (in many cases no Table of Contents information comes through the conversion process – and, even when some did, it had lost its numbering). Where the contents had to be created from scratch, each new content item created had to be linked to the specific HTML file to be displayed when that content item is selected.
  • Re-insert spacings in headings: The conversion process also loses the spacing in headings in the base material resulting in headings that look like this, ‘9.1Revised System’ instead of like this ‘9.1  Revised System’. To rectify this problem, the spacings have to be manually re-inserted into each HTML file of base material.
  • Compile the revised files into a single CHM file.

The first HLP file I tried this out on contained just a single Help document with some 130 pages. It took a bit of figuring out, but I eventually got the hang of it. However, the second HLP item was in fact made up of 86 separate HLP files all stitched together to present a unified Table of Contents in a single window in which the base material was also displayed. Many of these 86 separate files had 50 or more pages, and some had many more than that; and each page had to represented separately in the Table of Contents. It was a very long tortuous job converting all 86 HLP files and ensuring that each one had a correct Table of Contents (I didn’t attempt to re-introduce the spacing in the headings – that would have been a torture too far). However, that was not the end of it; the files then had to be stitched together in a single overall file that combined all the individual Tables of Content and that displayed all the base material. This involved inserting a heading for each document, in the master file; and inserting a linking command to call up the Table of Contents for that particular document. Oh, and I should also mention that the HTML Help File Workshop software was very prone to crashing – not a little irritating – I soon learnt to save regularly…..

This overall task must have taken at least 30 or 40 hours – but I did get there in the end. The new CHM file works fine and is perfectly usable, despite three of the documents being displayed in separate windows instead of the single main window (although I spent some time on this issue I was unable to eliminate the problem). Of course, the lack of spacing in the headers is immediately noticeable – but that’s just cosmetics!

No doubt there are specialists out there who would have made a quicker and better job of these conversion activities. However, if you can’t find such people or you haven’t got the money to throw at them, the experiences recounted above show that, with the help of the net, it’s worth having a go yourself at what you may consider to be your most difficult digital preservation challenges.

The Minimiser

Posted on by
Reply

I’ve been reading an increasing number of reports about how much time people are spending on their mobiles and of the many negative effects of such usage. Perhaps it’s time, therefore, for the emergence of a new breed of app explicitly designed to minimise one’s usage of the mobile. It would be capable of taking a whole variety of steps to reduce the amount of email you get; to summarise incoming communications for you; and to ask searching questions of you about new apps you want to load and new contacts you want to add. It would measure and report your usage of the mobile, and advise on ways that you can cut down the amount of time you are spending on it or reorganise your usage patterns so as to improve your quality of life.

Binding Sounds – Part 2

Posted on by
Reply

When I last wrote about creating the Sounds for Alexa book, I’d finished sewing the text block. The next step was to glue the mull (thin gauze) to the spine, then to glue some Kraft paper on down the spine on top of the mull, and finally to glue the blue and white end bands, as shown below.

Next, the colourful end papers I had selected were folded in two and glued with a thin, 3mm wide, line of PVA to the inside of the text block – one at the front and one at the back. Cardboard was then cut to the size of the text block plus a 4mm overlap all the way round, and glued to the mull and the tapes as shown in the picture below.

I’d elected to have a leather cover (as opposed to cloth) which is longer lasting and has a more luxurious appearance (see the picture below). However, the downside of leather is that it is thicker and less pliable and so it produces unsightly bulges when it is turned over the edges of the cardboard frame, and noticeable ridges at its edges. To solve this problem, the leather is pared down to a much reduced thickness at the points where it is to be turned and along its edges – in effect all but the central area as shown in the picture below .

Paring leather is a task easier said than done. Doing it manually involves using a very sharp blade and shaving off thin layers over and over again until the requisite thinness is achieved. This is a very messy process which produces lots of small fragments of leather which get everywhere. It took me a couple of hours to complete the task and the particular tool I was using caused me to lose the feeling in my thumbs and forefingers. No doubt it gets easier as one becomes more proficient, however, if I ever use leather again, I shall be investigating using a Paring Service which I’m told can be hired to do the same job much faster and to a better quality than I could hope to achieve, by using machines.

Once the paring was done, the leather was glued to the cardboard frame, the end papers were stuck down to the front and back boards, and the dust jacket was printed and cut to size and folded around the book. Finally a protective plastic cover was fitted to the dust jacket. The completed book is shown below.

Sounds for Alexa has now taken its place next to Alexa in our kitchen diner, as shown in the picture below. We will see whether it gets put to use as originally envisaged over the coming months.

U6.4 A summary view of the OFC future

Posted on by
Reply

Taking all the material from Units 6.1-6.3 into account, it looks like there will be a period of steady evolution before we start to encounter AI entities with the ability to do things autonomously. During this evolutionary period, the main individual applications that we use will become increasingly sophisticated and central to our lives; and we will make increasing use of applications and internet services which embed a degree of AI expertise and the ability to learn, and we will start to think they are normal and very useful. The amount of digital material we possess will continue to grow. It will become increasingly important to make arrangements for our digital accounts and possessions to be managed after we die. More and more physical objects will contain chips which we can interrogate and control through our computer systems. We will grow used to interacting with our computer systems by voice as well as by keyboard; and we will probably start to get used to virtual reality experiences.

At some point, the computer manufacturers will produce products in which our primary interaction with the system will be via a single AI entity. This will seem normal given what we have experienced before. The AI entity will take care of all maintenance, including backups, and will ensure that our files are always accessible and readable. As the AI entity becomes more knowledgeable, it will start to do more and more for us and we will have to provide less and less detailed instructions. The AI may start to see what we show it and know what it is looking at. There may be other AI in the house in other computers or in robots, and we will be able to interconnect them and instruct them to cooperate. While the digital world will increasingly be taken care of by our AI, we may start to value some of our physical possessions even more.

I can’t say I’m particularly confident that this vision of the future is what it will actually be like. Nor am I sure that it is of any particular relevance to any OFC project you are about to embark on. However, it was interesting to think through where things might be going. If there was any conclusion I would come to from this examination, it is that our digital world is going to be fully taken care of by an increasingly autonomous AI; and that, in the face of this, we should take increasing care of our precious physical possessions as they are the only things that are going to be truly under our control.

This is the last Unit in this OFC Online Tutorial.

To previous Unit                                    To Contents

U6.3 The future impact of AI

Posted on by
Reply

Artificial Intelligence (AI) is a term used to signify intelligent behaviour by machines – which really just means them doing more complex things than they have done before. Two significant milestones in the development of AI were when IBM’s Deep Thought programme beat Garry Kasparov, a reigning world chess champion, in 1997; and in 2016 when Google’s DeepMind AlphaGo programme beat a professional Go player. Current prominent AI work includes the development of driverless cars and trucks, and improving the ability of AI programmes to learn for themselves. It is generally thought that AI capabilities will continue to be developed for some time in just narrow areas of application, before eventually broadening their scope to become more general-purpose intelligent entities. Assuming this development trajectory, we can speculate that the way we deal with our digital objects and collections might be impacted by AI in the following series of steps, each one taking greater advantage of an increasingly capable technology:

A. AI to collect virtual objects at our specific request: The Facebook ‘on this day’ function that we can choose to turn on or off, is a good example of this in use in a contemporary system. In future systems we might imagine that we have an AI which is independent of any one system but which we could ask to collect specific objects across the systems we specify, for example, ‘collect all photos that we look at in our email, in Facebook and on Instagram’.

B. AI to collect digital objects at our general instruction: This is similar to step A except that we won’t have to specify the systems we want it to monitor. We‘ll just provide a blanket instruction such as ‘collect everything to do with any shopping I do’, or ‘collect all photos I look at’, and the AI will address the request across all the systems we use. At this stage the AI should also be taking care of all our backup requirements.

C. AI to understand what it sees in the digital objects: If we have asked the AI to collect objects for us, in this step it will be capable of fully understanding the content of the objects, and of having a conversation about what they are and the connections between them. At this point there will be no need for indexes to digital collections since the AI will know everything about the objects anyway; it will be able to sort and organise digital files and to retrieve anything we ask it for. The AI will also be handling all our digital preservation issues – it will just do any conversions that are necessary in the background to ensure that files are always readable.

D. AI to exploit our digital objects for us at our request: Now that the AI has control of all our objects and understands what they are, we may just be able to say things like, ‘assemble a book of photos of the whole of our family line and include whatever text you can find about each family member and have three copies printed and sent to me’.

E. Eventually we leave it all to AI and do nothing with digital objects ourselves: By this stage the AI will know what we like and don’t like and will be doing all our collecting and exploiting for us. We’ll just become consumers demanding general services and either complimenting or criticising the AI on what it does.

The last stage above reflects one of the possible futures described by Yuval Harari in his book ‘Homo Deus’ in which AI comes to know us better than we do ourselves, since it will fully understand the absolute state of the knowledge we have and be able to discount temporary influences such as having a bad day or some slanted political advertising. This clearly represents a rather extreme possible situation many decades hence; nevertheless, given what we know has happened to date, we would be foolish to discount either the rate or the content of possible development. However, we should also remain absolutely clear that it will be us, as individuals, that are deciding whether or not to take up each of the steps described above.

Throughout this period of the rise of AI, we will still be dealing with our physical world and our physical objects. AI may be able to see the physical world through lenses (it’s eyes), and be able to understand what it is seeing, and we may well get the AI to help us manage our physical objects in various ways. However, it won’t be able to physically manipulate our objects unless we introduce AI-imbued machines (robots for want of a better word). This too is a distinct possibility – especially since we are used to having machines in our houses (we’ve already made a start with robot vacuum cleaners and lawn mowers). However, having tried to think through the various stages that we would go through with using robots, I came to a bit of a brick wall. I found it very hard to envisage robots rooting round our cupboards, putting papers into folders, and climbing into the loft. It just seems unrealistic unless it was a fully fledged, super-intelligent, human-type robot – and that in itself brings with it all sorts of other practical and ethical questions which I’m not equipped to even speculate about. Perhaps all that can be said with any certainty about such a future of AI software and robots, is that humans will take advantage of whatever technology is on offer provided it suits them and they can afford it.

To previous Unit                                    To Contents                                     To next Unit