Tuesday, 31 December 2013

Prehistoric Park

I am being accused of over thinking things again - I wonder why.

There is a show on Watch right now called Prehistoric Park. I have to say that ,from a technical point of view, it is very well done. The plot is that they have a park/zoo that has a time portal device and they are going on safari to the past to collect animals from pre-history and bringing them to the park to preserve these extinct species. They have very well done CGI integrated with live action making it very convincing. It really is well done.

I have a slight issue with something that is total fiction being done as a documentary. The style of the show is exactly like the documentaries of real zoos, with ongoing updates on animals from the previous episodes, and the challenges they face with these animals, and so on. I can't help feeling that lots of kids, and probably even some adults, will assume it is real and that time portals exist! Maybe I am underestimating people. I have no problem with fiction, but making fiction in a style of a factual programme seems odd. Although War Of The Worlds original broadcast had the same issue :-)

But then I wondered about the plot. If you have a time portal then that changes the rules rather - you no longer have to consider any species as being extinct. It exists in the past, and the past is suddenly a place you can visit if you want. The past existence of these extinct species is not in threat (well, unless there are a lot of people with the time portals causing problems, but that becomes a slightly more fun plot line). So there really is no reason to preserve these extinct animals as they are already preserved in time. Indeed, taking them out of their normal environment is somewhat cruel and unnecessary.

But even if, somehow, it made sense to have samples of extinct creatures in a zoo, which may simply be a valid commercial venture in itself (if you had a time portal!), they had some odd ideas. They were deliberately getting animals from a time when they were close to extinction. Why? Why not go to a time when they were in their height when it would be easy to collect a couple of young specimens. In one episode they brought back a wounded wooly mammoth, depriving some hunters of their dinner. They had problems moving it as it was wounded, and then had to treat it. That is crazy, with all of time to choose from, make life easy for everyone by collecting young healthy animals that are more easily able to adapt to the change, easier to transport, and more likely to survive.

I suspect the answer is for me to avoid daytime TV.

Update: I see that even real documentaries can be works of fiction, from the BBC! (private eye).

Space dice!

So, we had fun yesterday playing an antique (looks like late 1930's) Monopoly game (thanks Mikey), and drinking some rather stong whisky (thanks Daryl and Guy). Kev won!

The game has original pieces and dice, but the dice were rather too worn to use. So I printed some dice on the 3D printer, as you do.

But just printing some dice is boring :-)

The design of a normal six sided cube die is pretty simple. I use slightly rounded edges, and put dimples for the numbers on each face in the usual way. I then arranged the printing so that it was done from one corner, with the aid of some snap off support legs to stop it falling over while printing. This meant that there was no top or bottom face (that always prints a bit differently). The lower faces were not quite as smooth as the top faces, but the dice produced were perfectly good. (I have not done hundreds of rolls to confirm no bias, yet)

But then I thought I would have some fun - being a 3D printer you can do things that would not normally be very easy, such as including a spherical cavity in the centre of each die.

I then dropped a smaller spherical magnet in to the cavity during printing. I had to put one on the outside to hold it in place. Continue printing, and now I have space dice! i.e. dice you could roll in space on to a metal plate and they would land and stick there. As the magnets are free to move there is no bias created.

The problem is that they stick together, so not easy to use as a pair in a shaker. Fine if rolled separately. But fun, none the less.

Anyway, for the game I did a pair of dice without magnets, but I did, for fun, put a small ball bearing in the centre cavity of each.

This created a rather unexpected, and useful, effect. The dice worked perfectly for playing, using the shaker and rolling on the board, but they had one rather useful feature. They did not roll far. I do not think we once had dice go bouncing off in to the distance under someone's chair, as is normal in any such games. They did roll a bit if you tried, but generally they just landed and stopped. Obviously in such cases using a proper shaker is a good idea to avoid someone setting up the dice and just dropping them, but it was rather interesting. I'll probably find someone has patented the idea already.

I also designed a set of loaded dice for fun.

Sunday, 29 December 2013

Most Reliable Broadband?

So, BT are advertising on TV that they have the "most reliable broadband*"

(* Compared to other up to 16 Mb/s providers)

This seems an odd claim for several reasons.

For a start, all of the other BT Wholesale based ISPs have the same underlying backhaul, so will be the same basic reliability. That means hundreds of other providers are just as reliable, making their claim a bit odd. I would not use "most" to refer to "equal first place with 100 others".

There can be a difference in provider though, some will be way more tenacious about fixing line faults, and oddly, A&A are known for that, even within BT.

But most importantly, as far as I know, BT do not do anything like A&A Office::1 package. This is two or three broadband lines, usually with different backhaul providers, and 3G backup configured to provide high reliability. It is provably more reliable than one broadband line. If BT don't offer that, but A&A do, then, well, who offers the most reliable broadband?

So, it seems, in my opinion, that BT's claim to offer the most reliable broadband may not be entirely true.

There are even ways to get more reliable broadband using multiple broadband providers and multiple endpoint hosted FireBricks tunnelling traffic.

I am not sure I can be bothered to hassle the ASA on this, as (a) they won't understand (they even think it is OK to call a copper pair "fibre"), and (b) the advert has already run on TV and the ASA typically just say "don't run that advert again". It is not like they ever make people actually do a retraction of the same scale as the original advert. Imagine that, BT having to do prime time adverts saying "not actually the most reliable broadband, sorry".

Saturday, 28 December 2013

Panasonic NN-CT562M fail

Well, it looks like this is a batch/model problem.

AO.com replaced the microwave for me, and arrived bang on time, very efficient. However, the replacement is also broken.

The previous one died totally on Christmas day. We did try the turning it off and back on trick, and even waiting a few minutes, but no joy, no display. Before the replacement arrived I doubled checked it was still dead, and it was not - it worked. So I tried the pop-corn for 3 minutes test. It failed.

However, to my shock, the replacement was the same.

I am worried AO may not believe me, or consider that I don't know how to drive a microwave, so this time I made a video.


But, obviously, I may be expecting too much. I don't think so, as I had the older model of this microwave, also 1000W, and 3 minutes was fine for pop-corn. The bag says 1½ to 2½ minutes for 950W. But lets be scientific here...

I used a cheap tesco own brand 650-700W microwave. The bag says 2½ to 3½ minutes. But 3 minutes created a full bag of nicely popped corn with no problem at all. I also made a video.


So, waiting for AO to reply to that.

Update: Got a helpful call - they agreed swapping to a different model would be sensible, and went through various models with me. It will arrive tomorrow. Very efficient - actually quite pleased with AO now.

Update: Got an Electrolux EMS30400OX, seems to work a treat.


Chess

I have not played chess in years, and am probably not that good. However, this will be something of a challenge, and I may have to see if I can find someone that can play.

Thanks Mikey.


Thursday, 26 December 2013

AO.com messing me about (result!)

After 3 minutes at full power in a 1000W microwave!
[See update at end]

So, got a new microwave from ao.com last month.

I have to say that, overall, they were very efficient and reliable delivering the microwave and delivering and installing the dishwasher we also ordered. Very impressed.

In spite of ticking the relevant box on the order form they did spam me trying to sell an extended warranty, but they shut up when I quoted the The Privacy and Electronic Communications (EC Directive) Regulations 2003 at them.

But the microwave was not quite right. We all thought so, but had not done scientific tests. The biggest clue was microwave pop-corn. In a 950W microwave this would usually be 1½ to 2½ minutes. This is a 1000W microwave, but after 4 minutes the bag was still flat! If you microwaved for 10 minutes it would just about cope, but clearly was not right.

I do worry about the safety of this - if this is pumping 1000W somewhere other than the food, where is it going.

By the time we had concluded that it was clearly not right it was nearly Christmas and over the timescale of The Consumer Protection (Distance Selling) Regulations 2000, so I decided that once Christmas was over I would contact AO.

Christmas eve it stopped working totally! I had to go find a microwave at 5pm on Christmas eve. There is, indeed, a Tesco value microwave on the shelf to take away in a shop that was open until 7pm on Christmas eve. Well done Tesco. It is not as big or as powerful nor have the grill and convention oven features or such, but it managed for Christmas day and Boxing day. So I am already inconvenienced and out of pocket. As it happens my son needs a microwave for his new flat, so it will go there eventually.

So, I contacted AO. Their reply was that they would contact Panasonic to inspect it and arrange a repair.

So I explained that I was requiring a replacement - that it was faulty from the start and that I was entitled to require a replacement, or failing that a refund.

Three times now they have stated the way they work is to inspect to decide if a repair or replacement is the course of action to take, so I have ended up having to quote 48B(1)b of Sale of Goods Act 1979 at them yet again and ask when they will arrange replacement.

They argued points in the Sale of Goods Act, suggesting that they get to choose if refund or replacement, and I argued points back at then.

In the end I pointed out that as it was faulty from new, and failed within 6 months, I was in a position to rescind the contract and demand a refund. If I do that I would be able to order a new one which, given their efficiency, I could get the very same day. So if they were not going to offer a replacement that would be similarly convenient, I would rescind the contract.

We'll see what happens, but I am concerned that they are not taking this seriously and considering it all under the optional extra manufacturers warranty rather than my statutory rights. I am, none the less, trying to be polite to them.

Update: After a lot of emails and a lot of quoting the Sale of Goods Act, and (one has to wonder) my blog post and tweet on the matter, I got a call. They are sending a new one for tomorrow and collecting the dead one, and giving a £20 credit. So, finally, we have a result. Well done, albeit a slight struggle.

Wednesday, 25 December 2013

Antiques

Well, Mikey has managed to find some interesting modern antiques for me for Christmas and birthday lately.

The original monopoly set was interesting, but the latest, today, was a genuine original Star Trek official tridimensional chess board set with all parts and even original packaging,

I don't really have anyone to play against, but it is starting to make something of a collection now, and my wife has kindly allowed me one shelf in a display cabinet. That is, indeed, an honour,.

LordK?

Well, apparently, I now own one square foot of a wood in Scotland (thanks, Louise). According to the web site, being a souvenir plot, it does not have to be registered in the Scottish land registry but is sold simply under a contract of sale for which I have the receipt/certificate.

The theory, it seems, is that Laird is a valid name prefix in Scottish for land owner and Lord is simply the translation to English, calling myself Lord would now be correct and valid, albeit misleading.

This seems more misleading in someways than Reverend as it is generally accepted (at least in UK) that one should not have a go at someone else's church or religion, so suggesting that ULC is, in any way, not a real religion, is not really allowed.

I'll ponder if I should use Lord as a prefix or not...


Update: Thanks for the various comments - it does appear to be, as I expected, a slight exaggeration of the truth, perhaps just short of a scam. It seems to be that I do own one square foot of Scottish woodland which seems it can be sold by simply contract (i.e. not registered). But the idea that owning that allows legitimate use of Laird or Lord is not quite sensible. The big clue is that they include a deed poll to change my name to have Lord on the front. If it legitimately allowed a prefix such as Lord (or even Laird), I would not need to change my name as such. Being Mr Lord Kennard or Rev Lord Kennard is a tad odd, and I'll not be doing that then.

I did email the land registry to find out if one can in fact register such a souvenir plot and what cost it would be. I suspect not a sensible cost, but would be amusing to do as presumably it means the registration for the area would need a small square hole removing from it.

Tuesday, 24 December 2013

3D maths objects


I have been having some fun with 3D objects created from mathematical formulae. A couple of good examples are the trefoil knot and the torus knot.

The starting point for both was wikipedia which has the 3D Cartesian parametric equations for this. Lets look at the trefoil knot first.

x = sin t +2 sin 2t
t = cos t - 2 cos 2t
z = - sin 3t

Using openscad I was able to make a simple module that does a translate based on this. But I found it looked better if I adjust the height (z) a bit, using h=4/3, and q=2/3.

translate([sin(t)+2*sin(2*t),cos(t)-2*cos(2*t),-h*sin(3*t)])
 sphere(q);

A simple way to use this was to make a loop, e.g. for(t=[0:1:359]) can in that I can make a hull of the function f(t) and f(t+1). This creates two spheres and joins them and does that for each step. This works, and creates a trefoil knot that is based on a circular cross section tube. It is slow, and you have to make the sphere have a lot of segments to appear round.

So I decided I could have more fun if I pushed a shape along the path. This means that I make a shape at each point, such as a triangle, square, pentagon, or even circle. The problem is that at each point I need to point the shape in the direction we are going. This means differentiating the original parametric function in three dimensions.

What I ended up with is :-

translate([sin(t)+2*sin(2*t),cos(t)-2*cos(2*t),-h*sin(3*t)])
rotate([0,90+atan2(1,-3*h*cos(3*t)),atan2(4*sin(2*t)-sin(t),cos(t)+4*cos(2*t))])
rotate([0,90,0])
rotate([0,0,($fn%2)?3*(t-30)/$fn:180/$fn])
cylinder(r1=q,r2=0,h=0.01);


The first part is the position based on t. Then I rotate to face the direction based on a differential of the original function, and then turn 90 degrees to face the direction. The then rotate the shape which is based on using $fn in openscad to make a triangle, square, pentagon, or so on. I actually rotate it so that it has a flat base on the bottom for easy printing but also, for odd numbered shapes, I rotate along the path to ensure all three points that touch the base are also flat as well as the top of the shape, and making it mobius so that the shape has a twist where it connects at the end. Finally a cylinder with small height and r2=0 is the best I could come up with to make a cross section to form the hull for each step. Using circle() did not work as it makes a cylinder with height 1.

The result is this. Using a pentagon ($fn=5).


The print looks pretty cool, and prints well on a Makerbot.

Having worked out how to do this, I created the same system for a torus knot. The formulae is actually very simple.

x = r cos pt
y = r sin pt
z = - sin qt

Where r is cos qt + o, and o is at least 2, but works well with 3. Basically o is the size of the torus. This all led to the openscad :-

translate([(cos(q*t)+o)*cos(p*t),(cos(q*t)+o)*sin(p*t),-sin(q*t)])
rotate([0,90-atan2(-q,-q*cos(q*t)),atan2(p*cos(p*t)*(cos(q*t)+o)-q*sin(p*t)*sin(q*t),-q*cos(p*t)*sin(q*t)-p*sin(p*t)*(cos(q*t)+o))])
rotate([0,90,0])
rotate([0,0,($fn%2)?q*(t-90/q)/$fn:180/$fn])
cylinder(r1=r,r2=0,h=0.01);


Which creates for $fn=5, p=3, q=7, o=3 :-


I have put on thingiverse, trefoil and torus knots.

But this does create a working principle for any sorts of 3D knot shapes.

Pardon!

So, the UK government finally pardon Alan Turning (BBC News) for his 1952 conviction for homosexuality.

This is just days after blocking access to LGBT web sites for most young people in the UK.

Monday, 23 December 2013

Parenting

OK, my youngest may be 18, but she has just told me something that happened at school some years ago. And I was so impressed I actually paid for the pizza for which she was nagging me.

She ended up in a church school. This was because that were the best local school for someone that is dyslexic, and they take on a number of statemented students regardless of religion.

I have to say that, at the time, I had my concerns over this, but what she has just told me has alleyed my fears.

At the start of year 9 they were all issued with a bible! She took it, looked at it, and put it in the bin in front of them.

Respect!

She was taught to make up her own mind on such things.

P.S. Pizza from dominos delivered by my son's girlfriend who promptly guilts me in to giving a tip even though the order is wrong - arrrrg!

P.P.S. This was almost as funny as discovering that my son's school called my wife to check his registered email address was correct, being an @fuck.me.uk address, and my wife confirming this and not telling me till much later.

P.P.P.S. I can see from comments my point was not that well made. I agree that she could (and should) have been more polite. However, as a child, standing up for your views when all around you (including authority figures) have a different view that they are forcing on you, takes some guts, and it is that for which I am impressed.

Weddings, Septic tanks, Women and Franking machines

I know my FaceBook persona is slightly odd, and it thinks I am only 18 or something, but still - I have "real" friends and relatives and make posts, and "like" stuff... So you'd expect FaceBook to have some clue when it comes to targeting adverts at me.

But what a fun selection: Weddings, Septic Tanks, Women, and Franking machines.

You have to wonder what FaceBook thinks of me (not that it thinks, but you know what I mean).

The scary thing is that, more and more, it is profiles like this that steer the way computers make decisions, and sometimes computer made decisions affect us in real life - from adverts to credit scores.

P.S. WTF have fireworks got to do with weddings?

Saturday, 21 December 2013

The perve list

One of the concerns that people have, quite sensibly, is that the new blocks imposed by many ISPs need specific requests to remove the blocks.

So, in order to have an unfiltered internet connection, you have to tell them you want the blocks removed. This is saying you want to access porn, or so it may seem. Indeed, some of the proposed legislation actually specified wording which meant that you specifically asked to access adult material. So, the ISPs now have a perve list. Understandably, people do not want to go on such a list.

As long as ISPs like A&A are allowed to operate without filtering, and people can choose such ISPs, we can avoid this.

Interestingly, when we did have a chat with ISPA members and some MPs a while ago, the MPs were suggesting that a choice not to offer filters would lose an ISP business, and several of us said that the opposite was true - that people find us out because we offer a clean, uncensored, internet service.

Well, tweets like this do kind of support our views on this.


Friday, 20 December 2013

Getting old

G on her GG
Very proud of my youngest daughter. 18 today.

But it does mean that they have all grown up now! If that doesn't make you feel old, I am not sure what will (well, I'll find out). It does make a bit of a milestone though.

Looking forward to a fun Christmas.

Happy Birthday G.

P.S. Now there are no kids in the house, please please Sky can you finally stop asking me for a damn PIN for shows before the watershed...

Thursday, 19 December 2013

46 seconds of fame

As I hope some of you saw, I was on BBC Newsnight last night (18th), talking briefly on the topic of network based porn filters.

I thought it would be useful to explain a bit about being on TV as it was all somewhat new to me, and not something any of us ever expect to happen.

It all started a couple of months ago. Being a little outspoken on some issues, including a few controversial ones, I was contacted by Sky News and asked to do a live interview. I was, of course, honoured, excited, nervous, a bit scared even. I had a couple of days notice, and this meant some sleepless nights worrying about what I would be asked (I had no more information than the general topic). We rehearsed it in the office, with other staff grilling me so I can try and be ready for anything.

With a 24/7 live news channel it is clearly very production line the way they do things. They wanted me there 10 minutes before I was on air, which seemed cutting it fine to me. On the day I got a train and they arranged a taxi to meet me at the station. Little did I know the station had a North and South exit, and you can guess what happened. Eventually the taxi and I found each other and I arrived with maybe 2 minutes to spare - a quick bit of make up, mic'd up and sat at the table with the interviewer and one other person. I guess they do two people in case one does not make it as that allows the slot to go ahead anyway. I was asked questions, and was on air for maybe 5 minutes, and then out of there. Yes, the taxi messed up so I had to wait half an hour, during which I was texted, emailed, and tweeted by all my friends and many customers commenting on my performance (thank you all).

The next time (on Sky again) was easier as I knew what to expect. This time it was a studio in central London and I was interviewed by remote. The taxi messed up at Waterloo, and I ended up walking in to the studio while the commercials were on before I was live. Sat down in front of a camera and a monitor, asked questions live for a few minutes, and gone (taxi waiting this time).

With both of these it is very difficult to get across what I felt were the important points. You are very much stuck with what questions are asked, and don't see a script or anything. Even though live, it is hard to put over the message you might have. In both cases I think I managed at least one key point, which was good. If I have to do any more, then I will know what to expect and I am sure it will all be very run of the mill stuff.

However, BBC Newsnight was very different. This is not down to BBC vs Sky, but down to the fact that it was a pre-recorded interview as part of a journalistic piece on the live program. Pre-recorded is very different, as I know from decades ago where I did live on-screen timing for sports events. Live TV is stressful for x minutes, and then it is over. Recorded is lots of faffing about until everyone is happy, and then they go off and edit stuff, so the work is far from over when filming stops.

I was quite pleased that the BBC decided to come to us. Apart from the chance to get everyone in the office madly making sure everything was tidy (well done guys), it saved travelling and annoying taxis. The interviewer, producer, and camera man turned up (no make-up, damn!). They had lights and camera, and we filmed lots of stuff, and in many cases several times over. Again, no script - they had ideas of what questions they wanted and I answered, often in different ways each time. They chatted over the issues anyway the whole time and recorded that on audio but off camera. I am sure this helped them understand the issues and create the overall piece that went out on air as they made some of the points and produced quite a good piece overall. Even do, trying to get my messages across was hard.

The camera man knew his stuff, and was clearly keen to do a few arty shots. He started by making the office less tidy(!), putting some boxes of cables on the floor and doing a close-up on a dangling cable. Why he did my shoes (sandals) I don't know :-) and he was very keen on the orc. We moved the orc in to the main office specially (losing an ear in the process) and I swear the orc got more airtime than did I! We managed to make the office look busy - they were later than planned but a couple of people were still in and a couple more came back in specially along with a friend (thanks Mandy).

Sadly, what went out was a total of 46 seconds, with one question and two of my answers put together. It makes a point, and gets A&A on the TV, which is all good, but it seems a shame spending well over 2 hours recording for 46 second. That is always the way with these things though.

Hopefully I'll be invited back, by BBC, Sky, and others. I think we do have important points to make from time to time, and I'm starting to get the hang of this TV stuff now. No autographs!

Wednesday, 18 December 2013

Mega

Thank you all, nice milestone for my blog.

Suffering for his art?

Poor old Vincent van Orc lost an ear and gashed his face last night during filming. We'll be lovingly gluing him back together today, but do watch out for him on tonight's BBC Newsnight as, hopefully, he'll be making an appearance.


Tuesday, 17 December 2013

The Internet Dark Age?

An anonymous article "The Internet Dark Age" published on Cryptome (here) alleges that BT and others have secretly installed back doors in modems in customer premises.

Unfortunately these paranoid ravings are causing a bit of a stir, with lots of our customers asking about this. We have had to post replies to this, and there is now an article in ISP review as well.

You really don't have to read much of the article to realise how crazy it is. It seems based on the fact that BT FTTC modems have a separate management VLAN, and that BT have chosen to put the management on a block of IPv4 addresses under 30.x.x.x. This is a non routed private network block that belongs to the US DoD which makes it ideal for a private network management LAN for BT. It does not mean it is connected to the US DoD either as a network or in any other way. It is just the block BT happened to pick for management.

The article links this to US DoD and to Snowden leaks suggesting that these modems are an integral part of the snooping being done by UK and US intelligence.

There is no actual evidence of any such links or indeed anything to suggest that this is any more than a management LAN which allows BT to do remote testing for fault diagnosis and upgrade firmware when needed. It is nothing special.

So what the hell is the point of such paranoid ravings? Who knows? They do serve to annoy people and also to undermine any real research or findings that happen later, I suppose.

At the end of the day no equipment that you do not control yourself (and even some that you do) can be 100% trusted. You cannot trust BT not to upgrade their modems to do snooping and filtering and man-in-the-middle attacks, but then you cannot trust them not to do that at the DSLAM or BRAS or anywhere else in their network. You cannot trust ISPs or transit providers not to do the same. You cannot trust your o/s provider not to do the same, or your TV manufacturer or indeed anyone that can upgrade firmware in anything on your network.

This is why we use encryption to access banks and many other sites, and can use end to end encryption for email and other applications. This is why we have firewalls. It is healthy not to trust anyone or anything when designing network security.

But there is no evidence of any such snooping. There is clear evidence against any sort of interception (e.g. man-in-the-middle attacks) because we, and our customers, can easily see what goes in and out of both ends of the back-haul. We have seen cases of bugs, for example, which caused packets to be dropped or changed, and these were picked up. Any attempt to change https traffic would be picked up very quickly and proved, so there really is no point in a major telco risking that happening.

Page 47 is a good read - the idea that the modem is a "white box" to throw you off (not being a "black box"), and the idea that it has a special hard-to-unplug RJ11 connector. Unbelievable.

I would have said the whole article is a joke, but I can't find the punch line. Are there really people quite that paranoid?

Friday, 13 December 2013

Ethics

One of these days I'll go on a course on ethics, but I imagine it will be a challenge for all concerned.

Having easily made a 3D print of an ABS key, I have an openscad file that will allow anyone with a makerbot or similar to make an ABS key quickly and easily, and do so from something as simple as a photo of a key.

It took maybe an hour to make the file, using some simple digital callipers on one of the keys, and only took that long because I am not really an expert on openscad yet or that used to making models. The mistakes I made were simple ones that someone more experienced would not have (adjusting size by 0.2mm to allow for plastic over print).

I was pondering posting the openscad file on thingiverse. And why not?

After all, the underlying issue here, that anyone could easily copy almost any physical key using a 3D printer, is not my doing. Publishing the openscad file makes copying such keys just a tiny bit easier. It is far more likely that someone will use the file for fun, or to make a spare key for their own lock, than any thief will use the file for some nefarious purpose.

I would not be publishing anything that is hidden or secret - the dimensions of the key are apparent to anyone with a key.

So is there an ethical question on this? Would it be wrong to post the openscad file for these keys? and if so, why, exactly?

I am sure people will have views on this...

ABS lock vs 3D printer

ABS make some nice kite mark 3 star rated euro profile locks.

These have a lot of security measures that make it very hard to pick the lock, or force it, or break it. They are almost impossible for a would be thief to get past, which is good news.

However, part of the security is the keys themselves. The keys are impressive, high precision flat keys that even include a magnet as part of the mechanism. You can't just walk in to a lock smith and get a replacement key cut!

For example this key (which is not my door key):-



However, these are no match for a 3D printer. Based on nothing more than a picture of the quality of the one shown above, an off the shelf 3D printer, and a magnet that costs 10p (bag of 50 for £5 next day from the internet), I can make a key that works.

A little bit of openscad, and a simple array for that key of no more complexity than d=[3,0,-1,2,1];
a=[-1,0,0,0,-1];

And we have a key in under 10 minutes!

Just add a small magnet (which took another 10 minutes for the glue to dry), and presto.

And it works! (video).

A few more details on how hard this really was. I measured the key this morning and made the openscad for it, about 30 lines of script. I printed, and it was a bit tight as the plastic tends to over print a fraction, so I adjusted the thickness 0.2mm. That fitted in the lock smoothly, so I then tried to glue the magnet in - I had to adjust the hole for that similarly and printed a new key. I found superglue was useless and managed to lose a magnet in the lock which I was able to fish out. One more time, and araldite for the magnet, and the key worked. I did not have to adjust my guess at the milling depths or positions at all! The most complex bit was the angled pointy bit at the end. The whole process took slightly longer as I had to print off a different key that was not actually my front door key for the pictures above - I am not quite that daft, honest.

Now I have a template I can easily print any more of these in around 10 minutes flat, and the levels are course enough that a simple picture like that shown above, or a modelling clay impression would easily be good enough.

This is not a slight on ABS - the vast majority of the security in these locks is the design that makes it hard to pick, force, jog, or break. Any mechanical lock will suffer that the key can be copied. The problem is that key copying used to need more specialised and harder to come by tools and key blanks, now something as simple as a 3D printer will suffice and no real skill needed. The time has come for proper keys with RF and encrypted codes - something that you cannot simply copy.

Thursday, 12 December 2013

Odd Christmas present

Well, I did wonder if this was the Christmas present to send to our router suppliers. For some reason they include a not-quite-cat5 cable in the box that has only 2 pairs, so will only work on 10/100Mb/s Ethernet.

These cables are such a pain in the arse that we remove them and replace them with a proper cat5 patch lead before sending the router out. Yes, the router will not do more than 100Mb/s, but the cables end up on other kit if you are not careful and cause much cursing and annoyance which we would not want to impose on any of our customers.

But what to do with a box of useless cables? One plan was to send to the router suppliers. That may be seen as a tad rude. In actual fact it is a "guess the number of cables" competition in the office, before they go for recycling. I think only staff can join in though, sorry :-)

"Why is it that all true geeks end up straying onto activities that were they not geeks would be considered criminal?"

I posted a video of this on FaceBook, and my friend commented as per the subject. Not that key cutting is criminal, but I get what he is saying. It's the challenge!

Yes, you can simply print a typical key on a 3D printer in 10 minutes and it does actually work a work!

To my surprise it does not even wear out that quickly - we have been playing around with this key quite a bit. Obviously nothing like as durable as a metal key.

I decided to design it from scratch as that is more fun, and it took me around 5 attempts to get it right. One of the key things was which way to print the key. As the print is built up in layers this matters. Printing it "up right", i.e. the bit you hold at the bottom, created a key that simply snapped when turned as that puts force between the printed layers. Printing horizontally worked a lot better.

I may have a go as higher security keys later. Just for fun.


Winding down?

Is it just me, or does it feel like it would be nice to take the rest of the month off? Not that I don't have enough leave left to do that, being the boss I can do what I like, but it does feel irresponsible. I do have lots of work I should be doing...

I am sure that 5 minutes of daytime TV and I'll change my mind.

Just seems one of those days somehow.

Monday, 9 December 2013

SIP2SIM relaunch?

We used to offer this, but using Three, 3G only, UK only. It was canned by Three, shame. But the good news is we look like we may have a new offering.

It allows us to offer a SIM that works like a SIP device. No, it is not an app, and will work on any old GSM 2G or 3G or 4G phone, but the calls are passed each way as SIP to/from a SIP gateway/service as a registered device. This could be a mobile connected to a company or home SIP system (asterisk, freeswitch, FireBrick, whatever) or simply registered on a VoIP provider platform (including A&As VoIP).

The good news is that this is not quite the same as before. It works on O2, but will have roaming even in the UK, as well as outside the UK. Roaming in the UK is pretty unusual, but I have already used the test SIMs on vodafone and EE. It costs more, and confirming the details of costs is one of the things on which we are working now.

We think it will be around £10 for the SIM and £4/month, with normal UK calls in or out on O2 at 2p/min. Prices to be confirmed, and plus VAT. Any VoIP calls charged by VoIP provider on top of that obviously.

Texting via http posts each way, or linking to an A&A number, with normal text charges on top of something for the mobile leg which we expect around the 2p to 3p per text. To be confirmed.

Data is currently NAT, but unfiltered, and not even via a UK link! But we are working on getting this via an APN that will allow us to do L2TP like the data SIMs. No promises on that one.

Obviously it will be easy to have a simple landline number on a SIM using A&A VoIP with it, but the real killer is when you link in to a PABX. See this Facebook video for an example of what I am talking about here.

Assuming we go ahead with this we are looking at SIMs in January. I'll be working initially as a trial until we have documented the limitations of how it works exactly.

Update: Yes, I changed "will" to "may" for good reason - watch this space.
Update: This really is touch and go now - we'll see what happens.

Update: Not happening, at least for now - lets see how it goes next year.

Update: OK, watch this space. We are still working on this. Just taking longer than we expected.

Update: Very close to launching now.

Interesting approach to SFI charges

XLN have taken an interesting approach to SFI charges, offering a sort of insurance.

Obviously, I can see why they are doing this. SFI charges are a lot of money and not something the ISP can really cover within the low profit margins, but also not something customers will want to stomach.

But the underlying issue is that BT are blaming customer equipment when they simply have failed to find the cause of a fault.

In principle, I am more than happy if BT were to prove the fault was end user equipment to then charge for wasting their time, but in our experience that is rarely the case.

So the idea of offering an insurance against the risk of an SFI charge is a novel approach. I hope XLN have the appropriate FSA stuff to offer insurance and are charging insurance premium tax correctly and so on. Or is it not insurance? I wonder.

The only thing I would disagree with is that they seem to be forcing this on their customers rather than simply offering it as an option.

I suppose I should work out if we should offer something like this at A&A. Comments welcome.

Saturday, 7 December 2013

§

The key codes on a keyboard have always included some keys that most people would rarely use. That is really not a huge surprise.

But even keys like ~ and ` have a use in things like csh. I use them on most days.

Even keys like & and % and \. I have few keys I have no use for.

But really, WTF is the "§" for? I have never, ever, ever, used § apart from this blog post!

And ± is the other part of that key, but even that I struggle to use. I was rather puzzled but the use of ± in the Netherlands on things like train times. They would have ± and a time. But they used it in a context of "approximately", like "~", so a train may be ±5 minutes late, which makes no sense. If really ±5min it would be either early 5 minutes or late 5 minutes, or maybe between 5 minutes early and 5 minutes late. They would use ±5 as meaning +5±1 or ~5min late, i.e. late, by 5 minutes or around that. Very wrong use of ±.

But I never use ± or § so WTF is it wasting a key on my keyboard?

P.S. http://en.wikipedia.org/wiki/Section_sign says it is a UK specific keyboard symbol, why?!

Thursday, 5 December 2013

Customer Service and ADR

Rather annoyingly I am losing sleep over the whole ADR thing again, but this time it is hypothetical debate by email with CISAS and not a case. This does, at least, mean we can sort ground rules out before getting a case.

The issue has come up over the types of dispute they can handle, which includes "billing", "service" and "customer service". The latter includes things like being rude to a customer on the phone. The debate started when I asked how they decide customer service issues and what they are, and they used the phrase "duty of care". I asked where it came from and they reference the implied term to use "reasonable skill and care" in providing a service. They have now said that this reference to the sale of goods act comes from OFCOM, which is a slight surprise, but does not change things.

I am surprised that providing a service (e.g. broadband) with reasonable skill and care could be taken as having to provide a level of "customer service". It seems a stretch to me, but that is, apparently, OFCOMs interpretation, so we'll work within that rather than arguing it.

Just to be clear - we want to provide good customer service. We understand that billing issues need correcting, and that a failure to provide service to the agreed level deserves the agreed level of compensation, but "customer service" comes down to keeping customers, getting referrals and getting good reviews. It is, in my view, not acceptable that someone can say we were rude to them and demand hundreds of pounds in compensation else blackmail us by threatening ADR which would cost us hundreds even if we "win". Let us be judged by our customers and their actions, not by CISAS.

It seems to me that they have several issues regarding "customer service" complaints and how they resolve them.

1. The Comms Act allows a complaint from someone who is not yet a customer, as it redefines customer to include someone simply seeking a service. CISAS agree that such complaints could only be about "customer service" as no bills or services yet. But the reference for the "duty of care" to provide "customer service" is an implied contract term, and there is no contract yet, so no duty of care. Ooops.

2. Even when someone is a customer, this duty of care comes from an implied term which OFCOM interpret in a particular way. However the sale of goods act allows such implied terms to be changed or removed by explicit terms in the agreement. So we do. We make it clear that we'll use reasonable skill and care to provide the actual service, but that "customer service" is not part of the contract. To ensure compliance with unfair contract terms legislation we even draw this to the attention of the customer when ordering. CISAS have now conceded that we can do this in the contract, but have not said what they would do about it (e.g. not take on customer service cases).

But then there is another issue which CISAS are still struggling to get their head around. I am still debating this one. It is the fact that they resolve disputes.

3. As a simple matter of policy, we agree customer service complaints. If someone says we were rude, we agree, and apologise. If they say they had poor customer service, we agree, and apologise. There is no disagreement, no difference of opinion, no dispute to resolve.

CISAS talk of whether or not our response to a customer service complaint "resolves the dispute to the customer's satisfaction" and cannot seem to grasp the concept that in such cases the complaint is not actually a "dispute" in the first place, and so the question of whether or how it is "resolved" does not come up. Also, being not a dispute, it is not a question of whether the dispute is within CISAS scope, or even how that is decided. If not a dispute, then there is no decision to make.

We're waiting for them to understand the concept of a "non dispute" and the fact that a "Dispute Resolution" service is not relevant to a non-dispute.

So, I'll keep you posted. At least we have some progress that the reference for customer service is apparently defined by OFCOM as an implied contract term which can be explicitly removed from the contract. That alone is massive progress. I wonder if other ISPs will follow suit on that. As I say, I am more than happy for our customers to judge our customer service in reviews and recommendations, but not via ADR.

Wednesday, 4 December 2013

Ultimaker rip off

So, I ordered an ultimaker, and even picked the express delivery.

They charged my card.

Then they tell me they have none, and won't for some weeks.

Then they tell me it will be even more weeks.

Eventually I cancel the order, and they refund, but the result is that I am £140 out of pocket due to card fees and exchange rates - a card fee even on the refund FFS.

They refuse to compensate me.

I may be investigating the logistics of suing someone in The Netherlands.

Tuesday, 3 December 2013

Wires only FTTC

BT have said they will be launching wires only FTTC next year. It will be a bit cheaper than the engineer install, and means end users getting their own VDSL modem. On face value, this seems good.

However, this is going to be a problem in many ways. The big issue with ADSL is that it includes all of the grey areas of modem to modem communications over a copper pair. This means you cannot tell if a problem is the modem at the exchange, the copper pair in BT, the wiring in the home, or the modem at the customer premises. So BT can, and do, blame the end user modem as the cause of many problems. It means we have to replace the modem before we can really start shouting at BT to fix things, and even then they will still blame the customer equipment. It is the reason SFI charges were invented in the first place - charges that are more than many years profit on a typical ADSL line and take a lot of time to dispute every month.

With FTTC and the engineer install BT are providing an Ethernet service. I.e. it is a service at the Ethernet port on the BT VDSL modem. They can test up to and including the modem and even confirm if the modem can see the Ethernet port up or not.

It means any line issue, whether modem at either end or copper pair is always BT's responsibility to fix. FTTC faults are almost always "clear cut", BT see the same issue and get it fixed, and do not blame customer equipment or make a charge.

With wires only we open the whole can of worms that is SFI charges and blaming the customer equipment. It also does not fit with the underlying service being an Ethernet service. It is all very messy.

We have considered not offering this cheaper wires only install, largely because the saving will not be enough to cover buying a new VDSL modem, but also because of the fact that BT would no longer be responsible for the modem.

However, we can see a whole raft of issues. We'll need to keep careful records of which lines are engineer install and which are not. And, sadly, we have to do this even if we only offer engineer install as someone could migrate an FTTC service to us that was not engineer install. We have no idea if the checker or tools to BT will tell is whether it is or is not engineer install, and we obviously need to know when reporting faults.

We then have the issues with engineer installs where the end user happens to have changed the modem to one of their own. We have to know, and have to ask the end user to put the BT modem back for testing.

It is going to be a nightmare.

Monday, 2 December 2013

Direct Debits

I am sure it is not just me that gets annoyed. I used to refuse to do any DDs personally, but they are not so bad when you understand the rules.

What is really pissing me off is some of the suppliers to A&A, big companies, that simply do not follow the DD rules. It drives me mad. The basic rules really are very simple: give the agreed advance notice of collection stating amount and date, and collect that amount on (or immediately after) that date and not before. Basically. They are allowed to collect up to 3 working days late, but not early. They can give a "recurring" notice that an amount is collected every month, etc, or a schedule of advance collections. But whatever, they have to give advance notice of collection one way or another.

When A&A collect by DD, we put a lot of effort in to meet the rules, including being very careful of the wording and layout of the advance notice we send, and ensuring we always send one with the agreed notice period, etc. We log what notice we send, and delivery reports in case of any query later. We have systems to allow any queried DD to be cancelled up to 2 days before collection and to immediately and automatcally refund any DD that we find afterwards was more than it should have been (e.g. if a credit note was issued later). It works well. We get almost no disputes and they are mistakes when they happen.

But suppliers charging by DD is really annoying. Almost none of them get it right.

  • We have one supplier who don't mention DD on the bill, but still expect to collect tens of thousands by DD without saying what date.
  • We have one supplier that thinks it is fine to collect thousands of pounds the day before the notice said they would.
  • We have one supplier that send two bills each month for different services, one states DD collection and one does not - but they always collect whatever the balance is on their DD run later in the month regardless of what the notice actually said, and get upset if we pay the one that did not have a DD notice by BACS credit to them (as detailed on the back of that invoice as a valid and expected way to pay).

I am getting so fed up that we are now, as a matter of policy, reclaiming all incorrect DDs. This is the only way to get these people to follow the rules, and is entirely correct as per the rules. You can bounce a DD on the day (or maybe the day after), or you can later reclaim a DD. In either case you need a reason, such as disputing the notice given, etc. There are a list of valid reasons. The bank have to do an immediate refund, and Barclays seem to do that correctly. You don't have to cancel the instruction when you do that. You still owe the supplier what is validly due.

I just want suppliers to follow the simple DD rules, or failing that, if it is too hard for them, let me pay by BACS credit. We have a good system in place so we never pay anyone late and send BACS payments reliably with the right reference. Insisting on DD collection and then getting it wrong is just daft, as well as creating a huge potential liability for claw backs if they do it to everyone.

Grrr...

Sunday, 1 December 2013

Barclays same day replacement debit card

I do make the mistake of watching adverts on TV occasionally. Barclays advertise same day replacement debit cards at selected branches.

Firstly, this is good. No reason not to have all of the tech to produce replacement cards at some, or all, branches. Well done Barclays.

However, I am annoyed at the advert. It stresses the serious inconvenience of losing one's debit card. For a start, who has only one card, but lets assume you lose them all.

For the Barclays replacement card you will have to go to an open branch of your bank, and prove who you are.

If you can do that, whoever you bank with, you can extract some CASH from them. It is really useful stuff that works in every shop. If it happens to be the day you need to buy loads of stuff, then you can get the cash to do that.

It will almost certainly be quicker to get that cash than a replacement card from Barclays.

So, whilst their service is very good, it does not address the issue they highlight in the advert any better than any other bank.

And, ironically, my son lost his debit card recently. He left it in a Barclays branch! They called and he said he would come in, but when he did (next day) they had destroyed it, and ordered a new one by post. He did not have option of replacement same day and so was without a card for two days just when he needed one. He got out some cash, as it happens. So lose your card, but don't lose it actually in a Barclays branch or you are screwed!