nigeria's electoral commission has postponed the presidential poll by a week — hours before voting was due to begin. the commission chairman, ya kub mahmood says it wasn't feasible to proceed as scheduled for logistical reasons but the delay was necessary to ensure a free and fair vote. the us president donald trump has declared a national emergency in an attempt to bypass congress and secure funding for his mexican border wall. democrat leaders have described it as "a gross abuse of power" and announced an immediate investigation into the declaration, which they say violates the constitution. relatives of shamima begum, the pregnant teenagerfrom london who went to syria to join the islamic state group, have asked the government to help them bring her home. britain's home secretary, sajid javid, has made it clear he'll try to prevent the return of people who've supported terrorist organisations. newswatch is coming up in 10 minutes but first it's click. massive attack's teardrop plays.

it is pretty far out there but it really suits the music and what they're about i think. in his birmingham studio artist, harley davies, is painting a unique work that is much, much more than meets the eye. it's mind blowing to think that, when you consider how much data there must be out there. it's interesting for the future, i'd say. the artwork is the album cover of the hugely popular and influential mezzanine album by massive attack. and to celebrate its 20th anniversary, the band agreed to have this music encoded in dna, and then added to several spray paint cans. it means harley's painting will hold thousands of copies of the album, and to find out how i have come here to a lab in zurich, to meet one of the pioneers of using genetic codes to store data. and so here's the freezer where we keep the dna. right, ok, this is where the magic happens.

exactly so in here, it comes in an enormous box, all really cold. and you buy in dna. we buy in the dna. dr grass has encoded the music already to be sequenced into the dna. that work is done by one of several companies now offering genetic code to order. so they make the dna in the sequence that encodes for the album. so we have the sequence of a, c, t and g and so they take a and then the other c and the other t so that will encode for, i don't know, 0010 something like that. and then you have to make — because the album is much more than just a few zeros and ones — you have to make a lot of those dna sequences. so the whole album is distributed over the tubes so there is no particular order. it starts at the beginning and at the end and so every tube contains a million different short dna sequences and every sequence has

a number stored in it to tell us where it sits in the overall picture of the album. so inside this tube is effectively about the equivalent of one of the tracks on the album. and how much does this cost? 50 megabytes, $1000 per megabytes, that's about $15,000 to store the album. it is a lot but you only have to do that once and then you can make enormous amounts of copies of it, because one key advantage of dna, i think, over all storage technologies we have, that essentially for free — nearly for free — you can make billions of copies, because there's an enzyme we have in our body, we know from biology, thatjust copies dna and so we feed this tube to it and itjust makes enormous amounts of copies of the dna we have in that tube. oh, wow, 0k< so that — there's an opaque section at the end of this tube. in there, there's billions of very,

very small glass particles and in the glass particles we have encapsulated the dna. so we have directed glass to grow around the dna to make that white, opaque powder. very similar to what you know from amber. right, and the amber is protecting it from decaying for potentially millions of years. here for millions of years, our dna in the glass, for probably 1,000 years, it protects the album from decaying. so you can still hopefully paly it in 1,000 years. how many copies of the album are inside here then — just the one? no, so we put a million copies inside. even if you don't spray with the whole can, you certainly have a copy of the album in there. so harley's picture paints much more than a thousand words. it is not only the first album cover artwork to actually include the album, a painting the size could store enough data to hold every album, picture, photo, book and a recording, audio and film ever created in the history of mankind. so how can we read the information?

that is one fly in the amber at the moment. this machine takes 17 hours to do it. it has come down from about a week but still, imagine pressing play and waiting that long. so we may be several years away from dna being practicalfor storage but at least it will hang around for thousands of years and, in a format we will always recognise when we see it. 0r hear it. a few weeks ago we visited the amazon spheres, part of the tech giant's headquarters in seattle. we met the people behind its voice recognition tool, alexa, and saw what else we'll

be talking to soon. as the tech gets better, it could one day become the way that we interact with our devices. now, that prompted this question from a viewer: thanks, simon, that's a really good point and, yeah, apple homepod gives some control to deaf users through the use of a touchpad but nowhere near enough to really use it. google‘s assistant can control a smart device by typing requests on a smart phone and the captions feature is available on the versions of alexa with a screen — news, weather and climate can be activated with tap. all of this is quite basic.

so abhishek singh has decided to show them how it could be done. he's created a simple algorithm to do this... the camera sees what he signs and turns it into text that alexa can understand and respond to. it only does a few words but the point is to inspire the big companies into action. last month, google released a couple of new accessibility apps for deaf users who use its android devices and laura and click trainee, maddie, have been putting them to the test. i lost my hearing when i was seven and about a year after that i got a cochlear implant which has helped me a lot, but even now in certain situations i find it really hard to hear. so when i'm in noisy cafes, or at a dinner party, my hearing is not the best. we have deliberately come to a coffee shop where there's real everyday noise all around us to demonstrate these.

maddie here has been testing them in various different environments. we are going to start off with google live transcribe. and it does what it says instantly and simply creating a script of your conversation. it can do so in 70 languages and dialects, with quite impressive accuracy. yeah, it seems to do really, really well with people talking. with one or two people talking, it works really well. the further away you get from it the worse it gets and the closer you get the more accurate it becomes. the underlying technology is automatic speech recognition technology and what that is is that's a way of us taking all sorts of known speech from recordings and basically training algorithms on top of it so that it learns all the nuances, all of the contexts that we understand as people. we also have here google‘s sound amplifier app.

you need to attach a pair of headphones to the device and from there it can turn up the volume on different elements of what you are listening to. it may be the quieter background noise you want to make louder, while keeping the main sounds you are listening to at the time, which could be some music, at the same volume. now, how useful did you find this was, maddie? i thought it was quite cool that you could play music and still hear stuff from the outside world at the same time. the phone's microphone picks up the ambient sound and, from there, the machine learning and artificial intelligence isolate the elements. that could make it possible to, say, make speech louder and the sound of an air conditioning unit quieter. people with worse hearing, it would be much more useful. because it just boosts that noise around you. so when you're less comfortable with your hearing it gives you that little bit of security that you could have that little bit of extra volume. it took be a good half—an—hourjust

focusing on the app, playing with all the toggles, it builds into the settings on your phone, so it took me a while to find the right settings for me. these are what i would call accessibility first applications, in that we're not taking an existing product and making it more accessible, we're making, in both cases, the real world more accessible using these technologies that exist on a smartphone and in the cloud today. and if you want to read it, you would have to have it in front the big game changer was the first hearing aids with the ability to communicate with and iphone came out. that then opened up a lot of possibilities. because you've not just got the processing power of the hearing aid, you've got the processing power of the smart phone as well. apple added similar functionality to their airpods last year, with live listen allowing you to place your phone or ipad near the sound you want amplified. and now starkey, one

of the leading hearing aid brands, will be adding full detection and a virtual assistant to their device that already features live translation and activity tracking. a live translation feature is promised, along with activity tracking, and an app to host a whole lot of data. it looks at your constant communication with other people and therefore it's measuring how much social interaction you're having. and there are also sensors inside the hearing aid, so motion sensors inside the hearing aid, which are looking at how much motion you've got. there has been found to be a relationship between cognitive decline and hearing health. but when it comes to google‘s latest releases, even if they're not proving quite perfect yet, they do harness the power of the fiercest weapon most of us have on us all the time. that's it for the short click. thank

you very much for watching and we will see you soon. hello and welcome to the programme. attacked while covering a donald trump rally, the bbc reporter there tells us what happened. hours of celebrities on the red carpet, is it time bbc news stopped going back crazy? on monday, president trump travelled to el paso in texas, speaking to a rally held to campaign for the will with mexico. in the media area was bbc washington

correspondant gary 0'donahue he reported the following day on what happened during the speech. journalists are regularly shouted at and abused. but this time, one member of the crowd took matters into his hands. i have reached the lowest level in the history of our town. he pushes the camera violently from behind into my cameraman. then, as he's restrained, he continues to yell abuse the media. fortunately, my cameraman was not hurt and the man was taken away. later, the bbc wrote to the white house, asking them to review security arrangements for the media at rallies such as this. and the white house condemned violence against journalists.

they asked supporters to behave in a peaceful and respectful manner. to tell us more about what happened and about the challenges facing journalists in america at the moment, garyjoins us now from washington. we cannot tell from the footage what exactly happened but it was described as a violent, you were there, can you describe what happened? what happened was that there were three of us, myself, my producer and my cameraman and we were on the media riser, the platform built for the cameras to go on in front of the stage where the president was speaking. one member of the public managed to get onto the back of the riser, there was no security there as far as we could see, and they started to run along the level of the platform where we there and they pushed my cameraman was my camera into his body.