Hi Banger,

I'm pleased to hear you're seeing data now.

I'm also pleased to hear that you haven't had anything keylogged, or your bank account raided!

In seriousness, I've tried to make clear in the original post (containing some further context and some external links, without going TL;DR), that my company is legitimate and genuine in stature and intent with this project.

Thanks and regards
Kes

Hello again, RobertoS

Again, I would like to re-iterate that we are taking infrastructure-based measurements - the Agent does not take any information on what you are doing, or how you are doing it.

Everything it captures is completely benign, and that benign measurement data is then processed centrally by our Analytics Engine.

Our service enables a view of the complete digital supply chain between a user and services that are elsewhere. This means that rather than just concentrate on a single tail circuit link, we are also able to steer your attention towards anything else that may well be a source of impairment that is affecting your performance - which will include the LAN where the Agent is based. Again, to be clear, this is not any form of LAN navigation/mapping tool - it simply considers the path taken to get to an external target (of which for this project there are 8), nothing else. Our analytics can then bring that in to the results that show potential impairment areas.

In other words, it helps cater for the "all the lights are green, but performance is still bad" scenario.

More often than not, and as you would expect, where the overall experience of a technology service is perceived as poor or bad, the reason for that experience is not down to a single solitary point. Some of it may come from your LAN, some from your tail circuit, some from your provider's core, some from elsewhere.

This then contributes to what we, and Ofcom, are trying to achieve. How well is Digital Britain behaving, and performing?

I hope this puts rest to your thought about what we are capturing.

So is the metric opinion based, measurement based or a mixture of the too?

Also on the targets, do you take into account that some operators have CDN on their internal network, so visits to the external website may not reflect the streaming performance for example.

Also what TCP ports are used outbound - this information should be in a FAQ so those with router or PC firewalls can configure.

"How well is Digital Britain behaving"
And you wonder why people think you may be monitoring a connections activity.

Be interesting to know how you cope with say an Xbox One doing an automatic update on a LAN while your agent is running on a PC.

Hi MrSaffron

Rather than write a lengthy essay of a post, your queries are best answered by a section on our website which includes FAQ, How It Works, and Why It Works pages - I've included the links, below. Please be aware that these links go to our corporate website and not specifically the BBFix project, so other information is also contained within the same web section and the rest of the site that (for example) mentions pricing. One of the TBB users has found this previously and asked whether we therefore charge - the answer, as previously posted, is that for coporate/Enterprise setups we do. This is a specific project with Ofcom, however, so again I will re-iterate that there is no charge to consumers who choose to join us on the BBFix project.

As for CDN, yes, there are considerations there - and more importantly there are ways to ensure we target effectively. These have been considered with this project, but specific customer adjustments are seen within our corporate engagements.

FAQs - http://www.actual-experience.com/analytics/faqs

How It Works - http://www.actual-experience.com/analytics/howitworks (please note the actual Agent size varies depending upon what system it is being installed on)

Why It Works - http://www.actual-experience.com/analytics/whyitworks

Hope this helps.

Thanks
Kes

In reply to a post by MrSaffron:

"How well is Digital Britain behaving"
And you wonder why people think you may be monitoring a connections activity.

Be interesting to know how you cope with say an Xbox One doing an automatic update on a LAN while your agent is running on a PC.

We are taking infrastructure measurements that once processed allow us to present a view on how that infrastructure would behave should you choose to use it in certain ways - we are not monitoring or capturing anything that identifies how you are using your machine or broadband link, or what you are doing with it.

As for your XBONE example, for that period in time then yes, your experience of trying to utilise your link for other purposes may well be affected. It is likely that our service may identify indications of buffer delay and potentially loss on certain points within the chain, for example. This would translate (potentially) to the output showing that your experience would have deteriorated during the time of that mass download/update.

In reply to a post by Kezkay:

In seriousness, I've tried to make clear in the original post (containing some further context and some external links, without going TL;DR), that my company is legitimate and genuine in stature and intent with this project.

Thanks and regards
Kes

Kes,

The question not whether your Company is legit or not, but how do we KNOW it?

You must appreciate that the Internet is beset with people suggesting that they are Legit (well, they wouldn't say otherwise, would they?)!

I suggest, in this instance, the issue is OFCOM's problem as they haven't made publically clear that they are undertaking this exercise, nor have they indicated who is undertaking it for them!

I note that you suggest that we individually contact OFCOM, but that really is NOT the way to do this!!

Hence, please excuse my natural reticence!

In reply to a post by johnjburness:

In reply to a post by Kezkay:

In seriousness, I've tried to make clear in the original post (containing some further context and some external links, without going TL;DR), that my company is legitimate and genuine in stature and intent with this project.

Thanks and regards
Kes

Kes,

The question not whether your Company is legit or not, but how do we KNOW it?

You must appreciate that the Internet is beset with people suggesting that they are Legit (well, they wouldn't say otherwise, would they?)!

I suggest, in this instance, the issue is OFCOM's problem as they haven't made publically clear that they are undertaking this exercise, nor have they indicated who is undertaking it for them!

I note that you suggest that we individually contact OFCOM, but that really is NOT the way to do this!!

Hence, please excuse my natural reticence!

Hi John,

I can completely understand and appreciate the comments you've made on this point.

The answer is that yes, ideally, Ofcom would place something on their site to publicly confirm, rather than ask individuals who wish to check to contact them. This was something that we initially asked for.

In most cases this would not have been a problem, but being an independent regulator Ofcom do not identify the parties that they are working with on their site.

This was why a compromise had to be agreed - in that it was possible for individuals to confirm with Ofcom themselves, rather than a generic message be created and published on the web.

It hasn't helped me in trying to assure communities that we are what we say we are, but it's the best we could do in this instance!

I hope that this helps.

Thanks
Kes

You talk of certain points in the chain, how do you differentiate between WiFi issues on the local LAN, versus backhaul congestion in the BT Wholesale network?

It sounds like if an ISP has users who use their connection a lot (because connection is good) your testing will actually show a poor experience?

For example my connection looks bad at 10am http://www.thinkbroadband.com/ping/share/8dea1261079...
but that is because the connection was being used for downloads

Which did not really answer the questions I had.

Human experience - suggest that the human has some input opinion or otherwise. That is not clear.
Neither is how you identify where in the supply chain an issue is, a nice Network and Compute Infrastructure bubble looks good in a quick guide, but not what people who understand some technical aspects want. Remember you are selling yourself to people who may as part of their day job be running the actual internet infrastructure.

My view is that you are doing measurements that others have been doing for years before and still do now, but have provided a different analysis slant so that when presented to clients (e.g. Ofcom) the non-technical can enjoy nice infographics suggesting that Skype performs less well at certain times. Or in other words at peak times when lots of people use the Internet some services slow down or peform less well.