How I fixed an unusable BT Infinity Connection

I’d been having problems with BT Infinity for a few months. Every so often web pages would just hang, or streaming video would freeze. Oddly it would often coincide with a new certain devices connecting to the network. Frustratingly, my connecting to Wi-Fi with my work laptop (Dell Latitude, Windows 10) would cause the Internet to stop working for a good half an hour.

I phoned BT and found their helpline very unhelpful. When asked to unplug the router overnight, I asked why this was needed and was told ‘it’s technical’. Throughout the process I felt as though I was just being read a script and not being listened to.

Anyway, I had managed to get my work laptop connected and had hooked into my company’s VPN when all the devices in the house stopped working again. My phone, the Apple TV, Kindles – all except for my work laptop. How odd. When I disconnected from the VPN, it too started to not work.

This made me think to try using Google’ DNS instead of BT’s. To my surprise, the Internet started behaving like a 70Mb/sec Internet connection should for the first time in months. My next step was to log into the painfully slow BT HomeHub router to try and change its DNS settings at the network level, rather than for each and every device. It turns out BT have restricted that, because normal people can’t be trusted to change their DNS servers, it seems.

After doing some digging, I discovered that the Apple Airport Express base station I’d been using to extend the range of the wireless network could be used as a NAT bridge and in place of the BT Homehub. I found this helpful post on the BT Forums which I will recap here incase BT decide to shut down or move their forums.

In short, you plug in the Ethernet cable from the white BT Openreach box into the WAN port on the Airport. Your username can be anything (I’m sure BT don’t rely on this for actual authentication, that’s tied to your line) and your password is simply a space.

You can then use Google’s DNS servers in place of BTs. One extra thing I had to do was set IPv6 to be ‘local link only’ under ‘Internet Options’.



This setup allowed me to retire my BT HomeHub for good, and the Internet connection has been flawless ever since. My Kindle can even connect, which is saying something.

Oddly, the default IP range for the Airport’s DHCP server is Class A, which means your devices won’t have the usual ‘192.168.1.X’ scheme, but this can be changed if needed. You can change this under the Network > Network options tab, though there really is no need other than it being a more common practice.


On the demise of the iPad

Unrelated: iPad's do not land very gracefully.
Unrelated: iPads do not land very gracefully.


So it seems iPad sales have fallen for yet another quarter. As someone whose iPad is their favourite computer; on the one hand this surprises me – why wouldn’t everyone and anyone want one of these fabulously useful and fun gadgets? But on the other hand I can absolutely see why many people wouldn’t have a space for this relatively expensive, yet limited device in their lives.
The problem is that a tablet does some things really well: browsing the web, watching videos, taking notes, editing photos. It also does some things very poorly: you can’t import music purchased from someone other than Apple into your music library (iTunes Match) for example and the apps are in general baby versions of their desktop counterparts. Need to rotate, annotate and save a PDF? Create a PivotChart? Have two Word documents open at the same time? You’re out of luck.


Most people don’t create

The iPad’s big differentiator is that unlike an iPhone, it can actually be used to create content (By content I don’t mean social network updates!). Witness the suite of applications Apple provide for free or next to nothing; GarageBand, iMovie, Pages, Keynote and excellent apps like Pixelmator, which is the only image editing application I’ve ever been able to comprehend (I even managed to ‘photoshop’ someone who was in from one photo and place them into another, while making it look convincing). The problem is, most people don’t create content very often. Outside of work and school (where people obviously do), most people’s computing needs boil down to what is the quickest and most comfortable way to consume content. The iPhone’s success in business was dominated by it’s consumer success (the so called ‘consumerisation of IT’) but the iPad hasn’t followed this because for most people, the phone is simply ‘good enough’ to check Facebook, lookup that recipe or watch Netflix.

The phone is the best compromise, for now

Most people have learnt this over the last four years of iPad usage. The iPad isn’t better than a phone at the ‘phone’ things people do (Facebook, messaging, email) and it’s not better than a laptop at the ‘create’ things people do (with all the edge cases these entail). If you’re going to put down £400 or more on a new computer, why would you buy an iPad when you know it’s not going to replace your aging laptop and you’ll still need to replace that thing when it dies too. So most people buy new laptop instead, and that would seem like a smart decision to me.

Phone sales remain strong and this is because the phone is currently the best compromise for mobile computing. A small screen with lots of sensors and are useful while you’re out and about – camera, GPS, compasses etc. The iPad is a more enjoyable and productive device to use than a phone because of it’s larger screen size, but that also means it can’t replace a phone because nobody wants to carry a large bag on them at all times. It’s not an ‘always with you’ device. The phone is therefore the best compromise between having a nice big screen, and having something that is always on your person. Will this always be the case however? Once smartwatches are able to connect directly to cellular networks and don’t need to be tethered to a phone, will they be able to take on the role of ‘always with you’, for messaging, directions, checking headlines etc? If this happens then what do we need the phone for? It’s no longer necessary to have this compromise of a smaller screen. In this case, would people decide that a watch and an tablet (or laptop, if the iPad or its competitors haven’t improved its software yet) will take on the roll of web browsing, Netflix and Facebook, fulfilling the rest of their computing needs? I could quite see myself using just a watch and a tablet, if both devices progress in the right directions over the next few years.

A technologist’s take on the A329M changes

Recent changes to the A329M have been completed, to much disapproval from local residents, including myself. At peak times, what was a 15 minute journey from Winnersh to neighbouring Bracknell can take over twice as long as it once did.

What has changed?

The A329M was until recently a two lane ‘duel carriageway’ that ran all the way from Reading to Bracknell. The A329M was originally intended to directly link the M4 to the M3, but these plans were scaled back in the 1970’s, no doubt due to costs (and you can bet local residents probably weren’t too happy about another new road being built through their town either). What’s changed recently is that just after the Winnersh junction, the left-hand lane has been turned into dedicated slip road to enter the M4, with the first exit taking you onto the M4 towards London, and the latter taking you onto the M4 towards South Wales. No longer is the A329M a duel carriageway for its entire stretch.

This has caused two problems:

  1. Traffic jams for those who want to stay on the A329M and head towards Bracknell.
  2. Increased ‘last minute’ lane changes by drivers who are unfamiliar with the new layout, or decide to change lanes at the last possible moment, one assumes to avoid sitting in the traffic jam (see problem #1). Many feel this is dangerous.

But why was this change made in the first place? The reason is that trying to join the A329M Bracknell-bound from the M4 could frequently take a ludicrous amount of time, often over 20 minutes to travel a what can’t be much more than a quarter of a mile along the M4 > A329M slip road. I know this because in a previous life, I used to take this journey daily. I also know many other people who travel from Reading to Bracknell via the M4 who shared this experience. One day a friend suggested that instead of leaving the M4 and keeping right to go towards Bracknell and sitting in the traffic jam, I should instead take the left lane back towards Reading, traverse the A329M back up a junction, exit at Winnersh and get back on in the Bracknell direction. To my surprise, this was actually about 50% faster than sitting in the queue to get on to the Bracknell-bound A329M directly from the M4. I would often do this and take note of a distinct HGV or van, so as I cruised by on the A329M I could look over at the traffic jam on the slip road and see how far the queue had progressed. The HGV would usually only have made it halfway through the queue.

So it is clear to me that there was a problem to be solved. The Highways Agency were not having a jolly when they decided to make this change, they had to do something – if it was quicker to take a ~2 mile detour to Winnersh and back than take the slip road, something was clearly very wrong.

So rather than simply moaning at the Highways Agency, I wondered what could actually be done to solve the problem? How does one even measure that a problem like this is solved? As the quote goes “You can please some of the people all of the time, you can please all of the people some of the time, but you can’t please all of the people all of the time”. Until we all go to work in automated, computer-controlled flying drones, there is always going to be some congestion as everyone decides they want to get to work for exactly the same time as everybody else each day.

Please bare in mind I approach this as someone who is interested in problem solving, but with no expert knowledge in large construction projects (I build software for a living, not roads) – but I thought it would be a fun thought experiment if nothing else.

How to solve it?

Our first problem – the traffic jam. This would seem to be caused by the fact that the volume of traffic (that’s us, traffic is just another word for people, remember) is too great for the road. How could this be solved, while not reverting the M4 > A329M junction back to it’s untenable situation?

The problem is that people actually get into the right-hand lane too early. I understand why they do this; to stay safe, it’s polite, and because the road signs encourage drivers to do so. However, from a pure ‘volume of traffic’ point of view, if we could encourage people to use the both lanes right up until the last minute, and to merge in an organised and safe fashion, studies show this would reduce queues because it makes use of the full capacity of the road.

This brings us to our second problem – aren’t those last minute lane changers dangerous? Obviously if you’re reversing up a slip road then you’ve left it a little too late, that is dangerous. Surely there’s nothing wrong in theory with changing lanes at the last moment, as long as it’s a considered manoeuvre? In order to encourage this, I would suggest that instead of going to the M4, the left-hand lane ends up merging into the right-hand. This gives drivers in both lanes the responsibility to merge. At peak times a speed limit would need to be lowered to 40MPH (enforced by average speed cameras) to aid safe merging, and also to help drivers in the right lane refrain from powering straight into the slower moving traffic that’s entering from the M4 further down, braking hard and causing tailbacks as the breaking forms a wave that spreads along the rest of the road. They may be unpopular, but variable speed limits have used to good effect on M25.

In Summary

  • The left-hand lane would have two distinct exits (instead of trailing off to the M4).
  • The land-hand and would merge into the right-hand lane.
  • During peak times a variable speed limit of 40MPH would be enforced.

I’m sure that much greater minds at The Highways Agency have considered this (and I say that respectfully). There is sure to be a reason why this wouldn’t work. I’d love to know. I can’t imagine how difficult it must be to try and keep all of us road users happy. I hope the problem is addressed, but for now my real-world solution has been to set off on my journey 30 minutes earlier and avoid the rush.