Every so often there is a report that shows a broadband league table of nations. Normally the UK and US are not near the top and this causes some comment. As someone, in the UK, who recently had poor broadband speeds at home I think I have a particular perspective. It is easy to see poor broadband speeds as a consumer issue that doesn't really affect large enterprises, but it does.
​
For years we were stuck with slow speeds at home which whilst not terrible were not exactly fast. We were excluded from streaming HD media, from having multiple users have an acceptable experience and had slow downloads. Others, I know, have it much worse than we did. There are many who can't get ADSL based broadband at all regardless of speed. The issue is that ADSL, the technology used most in the UK, is affected by distance. The further away from the source you are, the slower speeds you get as the signal degrades down the (usually) copper cables.

Universal provision of broadband is not a quick or cheap problem to solve. The UK has mostly has Fibre-to-the-cabinet (FTTC) which basically shortens the distance copper is used. Because fibre isn't subject to signal degradation in the same way as copper this is a good solution for many. This doesn't work for all as some are still too far away from the cabinets that supplies their line or the solution is just too expensive to be economic for a particular cabinet.

There are alternative technologies to ADSL which can provide reasonable speeds but they come with their own issues. 4G mobile signals have limited range and isn't present in many of the same remote locations that can't get landline based broadband. Mobile broadband is generally more expensive and has more limited data allowances. Satellite has near universal reach but can be very expensive and generally has more limited data limits. The real killer for satellite through is latency, the time it takes for a packet of data to get from your PC to the satellite and back down to the server. Wikipedia says about half a second per roundtrip. Compare this to my home connection that can 'ping' a London based server in less than 20 ms.

So why is this an issue for enterprises? Well quite aside from issues affecting their staff who might be remote workers, end-users and customers use consumer broadband. Anyone who has ever done calculations for website or application responsiveness will tell you that it is the edge cases that cause the issues. This leads to either catering to the 'lowest common denominator' or ignoring sections of society or potential customers. Modern websites have video and rich graphics, these don't work well on slow connections and use up that precious data allowance. Latency is an even bigger issue especially for websites with lots of small bits of content that load separately. There are optimisations and things you can do but poor latency means a poor user experience.
​
So the state of consumer broadband is an issue to enterprises.

​Images: Sashkin/Shutterstock.com and Chanclo/Shutterstock.com

The word unlimited is misused in consumer technology. Unlimited data transfer from mobile networks rarely is and unlimited cloud storage usually doesn’t last long. Is the expectation of unlimited anything unrealistic and what should the industry do about it?

Since the original iPhone launch smartphone data usage has spiked and is now the main buying criteria for many of us selecting a plan. When the iPhone launched on O2 there was an unlimited data plan. If you look around the market in the UK today you’re hard pressed to find unlimited mobile data and the market in the US in similar. Three still have their ‘All you can eat’ data offerings but they seem to have deliberately avoided the work unlimited. Even here though there are restrictions on tethering.

Consumer cloud storage has also had its share of issues with the word unlimited. Bitcasa famously offered unlimited cloud storage but the offer ended amongst much angst for its users. Similarly Microsoft made a big splash when they offered an OneDrive unlimited plan only to withdraw it a year later. The latest casualty of the unlimited storage wars seems to be Amazon whose unlimited Cloud Drive seems to being withdrawn.

So is unlimited realistic? As a consumer looking at an unlimited offering it seems to make some kind of sense. The things I am buying that is ‘unlimited’ is not tangible, it is in the cloud or in the case of data transfer is transient. This though is where the new world of cloud and everything online collides with the reality of the underlying technology. Whilst to a consumer this is all a bit abstract to a provider there is real world technology underneath. Real world technology with capex costs, electricity and cooling costs as well as bandwidth costs to be paid to providers.

I think there are a couple of types of consumer that are attracted by unlimited. There are those that are simply looking to avoid worry about hitting the limits. I fall into this category especially for mobile data. The other category some call ‘abusers’ but I can’t go this far, how do you abuse something that is unlimited? You can’t. In the data storage world there is subreddit for DataHoarders with their euphemistic ‘Linux ISOs’ who upload Terabytes upon Terabyes of data. These consumers though are costing the providers real world money because of the underlying technology costs.

I am not sure there is an answer to this conundrum. Consumers who don’t want to worry about limits will always be joined by those in the second category (which I cannot bring myself to call abusers). Maybe an answer is to set large limits which allows the first category to not worry but limits the second. But to do this the uneducated consumer needs to understand what that limit means. Do your parents understand what a 10GB limit on their mobile data plan means or are they just looking to use their phone without worrying? Maybe the answer is more education. Or maybe we just need to stop using the word unlimited, especially when we don’t mean it.

​Image: Zentilia/Shutterstock.com

​
​There have been a lot of high profile data leaks and hacks recently. How would you know if your data, your login credentials or something even worse was out in the public domain?

There are a few sites where you can search for your information and see if you've been unlucky, the most famous being Have I been pwned. You can also sign up for alerts at these kind of sites so you don't have to constantly visit and search. Pastebin is one of the places where sites like the one above check for leaked details. Pastebin has long been a favourite place for Ne'er-do-wells to paste log-in credentials and other leaked data in 'pastes'. Pastebin themselves seem to be aware of this and have their own alert service that lets you monitor their pastes pretty much in real time. You need a paid account to do this but they are not expensive. They simply let you create a list of keywords and if a paste appears that contains the text they send you an email. You can use your email address, your name, your address or anything else you are worried might get out there. I've done this for a while and the amount of alerts I get is a bit scary - my namesakes seem to have their passwords compromised on a regular basis. The most common type are 'dumps' of email and password pairs from compromised websites.
​

Now using the built in Pastebin alerts is interesting but for those with a technical disposition there is also an API. There are a few reasons why you might want to build your own solution. Firstly the Pastebin alerts only sends you a link to the actual paste and quite often they get removed before you click on the link in your email. By building you own solution with the API you can save the whole text to look at later. You can also have more than the standard 15 keywords if you build it yourself. The DIY approach means you can customise the alert mechanism to do something other than send an email. You could send an SMS or use of the of the mobile push applications like Pushbullet or Pushover.
​
The best reason for me though was this is interesting technology to play with. To this end I spend a couple of hours writing a simple monitor in Python. It's not the most elegant code I've ever written but it works and is published here on Github. 

Image: Minerva Studio/Shutterstock.com