Demon UK ISP spying on del.icio.us traffic?

If you use www.demon.net in the UK beware that they either have a misconfigured DNS or they deliberately intercept *all* outgoing traffic to del.icio.us

Demon does use squid caching, but they should not have a DNS A record entry that resolves to one of their own hosts !

tracert del.icio.usTracing route to del.icio.us [193.195.3.33]over a maximum of 30 hops:

1     1 ms     1 ms     1 ms  10.0.0.1

2    28 ms    25 ms    22 ms  anchor-hg-1-lo100.router.demon.net [194.159.161.32]

3    22 ms    26 ms    22 ms  anchor-access-3-s154.router.demon.net [194.159.161.129]

4    23 ms    25 ms    25 ms  anchor-inside-3-g5-0-1.router.demon.net [194.159.161.66]

5    24 ms    22 ms    24 ms  193.195.3.33

Trace complete.

I noticed this because del.icio.us appeared to be down, but I had access to a lynx brower through a shell account on a server in New York and discovered the it was reachable from NY but not from London…

No, this isn’t a CDN network being used by del.icio.us, from RIPE -

inetnum:         193.195.3.32 - 193.195.3.47
netname:         DEMON-INT
descr:           Demon Internet
descr:           Anchorage House
descr:           web-blocking-proxy-network
descr:           http://www.demon.net/

You can see that the IP address that del.icio.us resolves to belongs to Demon.

Other websites, such as www.nytimes.com are not effected. I’d ring Demon and ask them but apparently everyone goes home at 5pm because that’s when the help desk closes and they don’t have an email address for the help desk either. Useless.

Why Austin Reed can do better

I’m constantly amazed by shop assistants in large London stores that cannot effectively interact with customers.

I know that in almost all cases, this isn’t the shop assistant’s fault, it’s the fault of management. Management who fail to realise that every customer interaction is important and that fail to put in place effective procedures for dealing with customers so that the outcome of every interaction is maximised for both the shop and the customer.

On visiting Austin Reed in Picadilly Circus, London, I asked an assistant where the changing rooms were and if they provided a tailoring service. I was told where a free changing room was located and that, yes, they did a tailoring service.

Now, I’m standing in front on the shop assistant with two pairs of trousers in my hands. Why do you think I’m asking about a tailoring service? Maybe because I want the trousers I’m holding in my hands taken up?

So, instead of answering simply ‘Yes’ in response to my question of ‘Do you do a tailoring Service?’ The shop assistant should have answered ‘Yes, if you’d like to try them on, then step out of the changing room, I’ll pin them up for you’. Instead I have to prompt the assistant ‘So do you just want me to fold them up to where I need them taken upto?’ the answer was non-committal, suggesting this simply wasn’t a question they were used to dealing with.

At Jaeger in Oxford Circus, the assistants actually ‘assist’ you. Jaeger’s shop assistants realise than men’s legs do not all come in short, regular, or long lengths. You try the trousers on, step out of the changing with your shoes on, then the assistant pins the trousers up to the length you would like them.

I didn’t bother to get the trousers taken up by Austin Reed, they just didn’t seem interested (I’ll get my dry cleaner to do it instead and save the tube fair and time of going back into central London).

This is a missed sales opportunity for Austin Reed, had they taken the trousers up for me I would have had to return to the shop to pick up the trousers, at which time I may have made more purchases.

In the same visit to the shop I also saw a potential customer disappointed. A man had mistakenly thought an item was in the sale, when it was not, as he had seen the item in the window adjacent to items that were on sale, although this item was not. I doubt mixing sale and non-sale items in a window display is some sort of sales tactic to get people through the door, because the response of the shop assistant was simply a shrug when the potential customer pointed out how confusing this was.

The shop assistant should have apologised for the confusion caused and offered a small discount. Subsequently the shop assistant should have given his manager feedback about mixing sale and non sale items in the window display.

I know the first of these two things didn’t happen as I watched the potential customer leave the store clearly annoyed, and I somehow really doubt the second thing happened.

Austin Reed, try harder, you can do better.

The Girls of justin.tv

On visiting justin.tv today I noticed a number of new channels. New channels appear on justin.tv infrequently and according to some internal business logic (at least I hope there is logic involved) that is never disclosed to viewers.

I note with interest that the new channels are all created by young ladies :-

• spgirl
• ashleymaria
• mimitv
• daikatana

Currently, broadcasting on justin.tv can only be done by obtaining a betacode. Therefore I can only assume that either the betacode waiting list of was chock full of young ladies simply desperate to start broadcasting, or the guys at justin.tv cherry picked from their waiting list to choose those channels that would attract the most viewers.

I’ll leave it to you to decide which is the most likely reason.

Intermittent variable reward in user interface design

I’ve been wondering why it is that some web sites with very poor usability still manage to retain users. Of course the first example that comes to mind is MySpace. Many people simply attribute its stickyness to the fact that people continue to use MySpace because their friends also use it, and I’m sure this is indeed the most important factor.

However my half baked theory is that a contributing factor maybe that the user interface design of MySpace creates a type of intermittent variable reward experience for the user. On almost every page, especially user profile pages, the user’s brain is forced to visually hunt for menu items, during which time the user’s brain is thinking, for example, “Where is the menu item to navigate to this users pictures?”, and then from a fraction of a second, to several seconds later, the brain is rewarded with the menu item, “Ahh, there it is!”.

This can be compared to the task of finding the best way to place a card when playing Solitaire, a trivial task requiring the visual scanning of the cards already placed, and a task with a small reward, but nonetheless one that is oddly addictive.

The intermittent component for a web site user is the time taken to find the link, and the reward is finding the link itself. Admittedly the variable component of the reward cannot readily be explained by the user interface design. This aspect has more to do with the content found after clicking the link.

However I find it amusing to think that tiny reward of a finding a link which occurs when links are hard to find, due to bad usability, is not sufficient to stop users from continuing to use a web site, when variably rewarding content is also offered.

Why justin.tv needs to try harder

Justin Kan, the founder of justin.tv, is a pioneer of live mobile video broadcasting on the Internet.

A walking advert for the possibilities of live mobile video, the justin.tv viewer is given access to a live video stream of Justin’s life 24 hours a day. Other justin.tv channels offer streams from a diverse range of people, some stream mobile video, some don’t, and few match the uptime of Justin Kan’s own stream. However the emphasis is on live video, and this emphasis defines a new niche in internet video broadcasting.

However, despite Justin’s valiant self-publicising efforts justin.tv’s first mover advantage is rapidly being eroded, and may already have disappeared with slick competitors like Ustream and Operator11 offering lifecasters similar features. The race to critical mass, with the prize being a sustainable business in this new niche space has now began.

In any closely fought race however it only takes one trip to leave a gap that can never be closed, and justin.tv is tripping over its own feet, while it’s competitors are not putting a foot wrong.

Slow to add features, and unresponsive to bug reports, justin.tv has done little to improve the usability of its web site in the last two months. But then, today, a new web site design was rolled out. During my morning coffee I logged into justin.tv to check out the archives of the prior evening’s east coast events and was surprised to see a redesign had taken place.

Surprised because there was no substantial advance communication of a major user interface change on Justin’s blog. Yes there is a post about the new design, made at 11.45pm and I understand from fellow chat room users the design was changed shortly thereafter. A few hours notice of deployment of a major redesign just smacks of a lack of advance planning. And when the blog post announcing the redesign is made, and the deployment of the design, occurs while you are asleep it is no notice at all. (N.B. I’m on London time).

What’s worse is that the redesign, while adding some features, and tweaking others, actually decreases the usability of the web site, and what’s more, it’s not pretty. And yes, pretty counts, just ask Apple and other companies that understand that people like beautiful things, and even more so, people like beautiful usable things.

Compare justin.tv to Ustream. Justin.tv is a web site designed by programmers. There is a reason why there is a whole industry based upon the design of web sites, and another whole industry based upon the writing of software. Programmers do not make good web designers.

Here is a list of design and usability issues :-

• The horizontal strip at the top of the page show pictures representing each channel. Unless you mouse over the picture you cannot see the channel name – global navigation must be near perfect which this is not.
• The same strip bleeds off the screen to the right, making you feel like you need to scroll your browser window.
• You embedded a big ugly ‘Live’ button that looks more like an RSS feed button into the already small video screen.
• This same button doesn’t scale well when you view the screen in full screen mode.
• When a stream goes offline, the ‘Live’ button stays on the screen.
• When you browse the archives you can no longer see the chat window. Annoying as a common use case is to chat while watching the archives while waiting for the stream to go live again.
• When browsing the archive you cannot easily see if the channel has gone live.
• When watching the archive video you have a huge button at the bottom of the small screen to open the calendar window – remove all superfluous buttons and watermarks from inside the video window please !

These are my observations, and other people’s from the chatrooms in each channel on justin.tv. It doesn’t matter than only myself and maybe half a dozen other people noted these issues in the time I spent on justin.tv this evening – the fact remains there should have been no issues whatsoever. This is sad given the amount of effort that must have gone into this redesign. Even more sad is that I would gladly have given my time, as would many other justin.tv users, I’m sure, to have beta tested the redesign before it went live.

The issues listed above are design and usability issues that are obvious to actual users of the website (as opposed to developers thinking like users) and could have been spotted and remedied before launch. I won’t even mention the actual bugs. This is another stumble in the race.

To be fair, progress is being made and the new live page is pretty neat. But the rate of progress is not fast enough, which as a user thinks Justin Kan is a gutsy pioneer, and enjoys spending 15-20 hours per week on justin.tv, and wants justin.tv to succeed, is frustrating.

Also worrying is that if the roll out of the new design is indicative of justin.tv’s management and execution abilities they won’t be a finisher in this race. First mover advantage, a bunch of smart and dedicated people, and some VC funding does not guarantee success, execution is also important.

In short, try harder justin.tv.

Why reading Digg is like eating celery

Finding interesting articles on Digg is like eating celery. Eating celery requires more calories to be expended than is provided by digesting it.

Reading through the massive volume of uninteresting, yet seemingly popular, articles on Digg is the same. It is frequently a zero sum game where the objective is to find the interesting article. And, on the occasion you do find an article that interests you, often you’ve spent precious minutes of your non-work time that could have been spent better, erm, working.

On the other hand reading Slashdot is like eating a small but juicy piece of steak. For the effort of eating the steak (providing you have a sharp knife!) you are rewarded with a lot of calories. Similarly for the effort of visiting Slashdot you are, generally, rewarded with some interesting story.

While some people enjoy celery, I do not. For this reason I only visit Digg when I’m starving.

Of course this viewpoint is a subjective one. If you like a diet of Ajax tutorials, Mac rumours and Photoshopped pictures you will find Digg to be a veritable smorgasboard of links to such things !

Extending the steak or celery analogy further we have social network web sites, MySpace and Facebook. Guess which is celery and which is steak !