Monthly Archives: January 2009

Epic Networking Fail: D-Link, 2Wire, Belkin, Netgear AT&T and Fry’s

It seemed like a simple problem.  My friend had a DirecTV DVR in the bedroom, and a PC in the kitchen.  She’d just loaded up DirecTV2PC, and wanted to stream HD from the DVR to a new Sony PC.

The PC was beefy, more than capable.  But the network was weak.  No problem I thought, just throw in some fast powerline networking and she’d be watching the DIY channel on her PC in a snap.

I’d been around technology too long to know that nothing is ever a snap, but I’m too much of an optimist.  A snap, alas, it was not.

So I picked up a pair of Belkin Powerline adapters that claim to deliver 85 megabits over standard AC power plugs.  The HD video probably topped out at 6-10 megabits, so it should have worked just fine.  So I thought.

Setup was easy.  But the results were dismal.  Stuttering and dropped frames made the stream unwatchable.   I should have asked Patrick Norton – he railed on powerline in a recent Tekzilla, and it appeared that I had the same problems.

Or was I?  Was it the network, the DVR, the beta version of DirecTVPC, or perhaps the PC’s video drivers.  I needed to isolate the problem.  So off I went to Fry’s (a local high-tech shopping bazaar) to pick up a 50 foot Ethernet cable, to isolate the network out. 

I was again reminded of the power of copper – and why I’d retrofitted my home with Cat6.  Wires solve problems.  Once I’d hooked the DVR directly to the switch, the video looked great.  Full on HD, no stuttering, no drop outs, no problems. 

Belkin Fail.

I wasn’t about to run twisted pair from the kitchen to the bedroom, though.  That would have been way too expensive, and involved wires, permits, electricians and more.  Instead, I trundled off to Fry’s to find a wireless solution.  Wireless N claims to moving HD signals around the house with ease, and I’m sure I’d find an N solution to bridge from the DVR to the PC.

Over at Fry’s, a nice clerk steered me towards the Netgear WNHDE111-100NAS, a wireless bridge that promised, according to the clerk and the box, to "Connect any Ethernet-enabled media or gaming device", to any "existing router/gateway."

I needed N on both sides, and the clerk pointed me to a D-Link DIR-628, a decently priced N router and hub that he insisted would easily connect to the Netgear bridge.

So I took them home, and the fun began.

My friend has a DSL router/modem combination made by 2Wire and sold by AT&T.  I was leery of having two routers, with NAT and DHCP daisychained together, so I attempted to put the 2Wire into modem mode, disabling the wireless and DHCP. 

Welcome to my first fail.  I could easily disable the wireless, but there was no obvious way to put it into modem mode.  A search of the internet, and 2Wire’s site failed to deliver guidance.  So I opted to install the D-Link router, and tweak settings.

2Wire Fail.

I ran D-Link’s nicely imagined installation wizard, which included an option to install along with an existing router.  Just what I needed!  I let the configuration program do its magic.

Black magic, apparently.  After an eternity of configuring, checking, setting and rechecking, the entire process failed.  Completely.  It’s advice was to shut the system down, and try again.  I did.  Another 15 minutes of configuring, checking, setting, etc, and it failed again. 

EPIC FAIL D-Link

It lifted my hopes so high, only to quash them miserably.

Still, after an hour of tweaking, I got things mostly working between the reticent 2 Wire and the lying pile of poop D-Link.   Time, finally, to set up the Netgear bridge.

I opened the box, and started reading the quick-start guide.  My hopes turned quickly to ashes, as I realized that the Netgear would work in only two modes – as a wireless bridge when paired up with another WNHDE111-100NAS, or when plugged in, WITH A PHYSICAL CABLE, to a router.  It reminded me of that old punchline, "If I could walk that way, I wouldnta needed the talcum powder."

If you’re counting, that’s two fails in one device.  First to Netgear for implying on the box that you could create a wireless bridge with just one box.  At least Belkin put two powerline-units in a single SKU.  And finally to Fry’s for putting such a brain-dead clerk out on the floor.  And you wonder why retail is dying.

EPIC FAIL NETGEAR

EPIC FAIL FRY’S

I made a half-hearted attempt to connect the Netgear to the D-Link using each device’s "Push ‘N’ Connect" WiFi Protected Setup (WPS) mode, but that, too failed miserably – neither device discovered the other, leaving each lonely, deaf, dumb and blind to the other.

FAIL NETGEAR

FAIL D-LINK

Sure, I could have gone back to Frys, returned everything, and picked up another Netgear Bridge.  But frankly, I’d lost patience with the whole thing, my friend said that she was planning on moving the PC within wired-up distance to the DVR later this year, and if I didn’t have a glass of wine immediately, someone at Fry’s was going to get hurt. 

So all that’s left, now, is to return everything to Fry’s when I’ve calmed down a bit. 

The moral of the story is clear.  First, 802.11N still isn’t ready for prime-time.  WPS might be sold as a panacea, but don’t count on it working across different vendors’ equipment.  And if you are considering getting a combo router and DSL or cable modem from your ISP, think again.  Your upgrade choices will likely be limited down the road.

WiFi and networking have been integral parts of our lives for more than ten years.  Yet it’s still too hard to get this crap to work together – and HD just makes it worse.  I still believe that wires, not waves, are the best way to go.

Web Video’s Powerful Third Dimension – Depth

image Traditional web and television advertising gets purchased mostly via a two dimensional graph of reach and frequency.  But internet video has added a third dimension to this analysis, creating a far more effective and efficient way to galvanize a core audience.

Most of my discussions with large brands and agencies over the past 15 years have focused on reach – or raw volume of eyeballs exposed to an ad unit, and frequency – how many times a particular set of eyeballs sees the message.  These two metrics pervade traditional television advertising, and result in a bombardment of messages.  If you don’t skip commercials via your DVR, and watch television with any frequency at all, chances are you see the same old, tired, dull ads over and over and over again. 

The truly bad ones become legendary – like Toyota’s "Saved By Zero" campaign that was widely lambasted – in part because they reached so many, so frequently, with such a terrible song.  Force feed enough pabulum to anyone frequently enough and they’ll  puke all over your message. And that’s exactly what happened here.  Toyota claimed that the campaign was successful, but I’m not so sure – they still swung to a quarterly loss, despite amazing reach and painful frequency.

On the web, reach and frequency are often abused to make up for the poor performance of most banner ads.  If they’re not screaming at us, or blinking madly, we mostly ignore their presence.  Thus the text web has evolved into mostly performance-based marketing – paying sites more and more on clicks and actions, and not based on reach and frequency of banner impressions

But just as the text web developed its own powerful and effective mechanism that’s supplanting reach and frequency, the video web has begun to discover its own unique selling proposition.  It’s a new factor – "Depth" – that adds a third dimension onto the advertising landscape by grafting a host’s social graph onto the 2D media landscape.  

Take a look at some of the best online video shows out there, including our own Tekzilla, Hak5 and Diggnation, but also Next New Networks’ Thread Banger and Non Society, along with You Suck at Photoshop, TikiBar and Wine Library TV.  Each are dominated by engaging, personable and talented hosts who have a direct, intimate connection with their audience – which is manifested by huge followings on twitter, Facebook, MySpace and elsewhere.

Their shows build on that direct connection by enabling an unparalleled Depth of engagement to a marketer’s message.  Whether it’s Kevin Rose extolling the virtues of Michelob, Veronica Belmont explaining the benefits of HP’s Media Smart Server, or Meghan, Mary and Julia expounding on the sexiness of Degree anti-perspirant, these shows provide a depth of engagement with a brand that simply isn’t possible on any other medium.

How powerful can this "Depth of Engagement" axis be for a brand?  We were part of three separate surveys in 2008 that proved the powerful and cost-effective benefit of this new metric.  Our own internal studies show amazing recall, awareness and action from our audience – 93% unaided recall of two or more sponsors, and a whopping 48% that purchased a product or service from one of our sponsors.  But it’s the two other – sadly proprietary — studies that really hit home.  In the first, a major marketer found that in-show host endorsements, along with in-depth post-roll video, provided an amazingly cost effective way to create awareness, affect intent, and drive positive brand impressions.  In the second, another consumer marketer credited our integrated plan as the linchpin for building a strong, core, passionate group of ambassadors that proved critical in successfully repositioning their brand.

The text web has been built on CPC.  But the video web is as different from the text web as it is from traditional television.  It’s a new medium and it provides marketers a new weapon in the battle to convert consumers into paying customers and brand ambassadors.  Just as the web gave rise to interactive marketing agencies that built core competence on SEM and CPC marketing, the video web will create a new discipline, and a new way to reach audiences by layering a third dimension onto the Reach and Frequency graph. 

Internet video isn’t the end-all for every campaign.  But brand managers, marketers and agencies need to add Depth into the mix, along with Reach, Frequency, SEM and CPC.  Those that don’t will be missing a new, cost effective way to reach core consumers – while their competitors will be building deep, and engaging conversations with their former customers.

 

See, you don’t need to lie about your web video numbers to make money!  Read my previous post for details about how Internet video is shooting itself in the foot by faking reach and frequency, instead of focusing on its unique characteristics.

The Lies Must Stop

liarI’ve been smoldering over this issue for the last 12 months, but I’ll remain silent no more. The lies about how web video views are calculated are simply, quite simply out of control – and threaten to kill our industry.

What sent me over the deep end? A second email from the producer of a new web series on entrepreneurship, looking to strike a deal with Revision3. Here’s the money quote:

We currently have a web show on REDACTED.com called REDACTED that is doing very well. In less than 2 months, our videos on their website generated 9,242,941 views.  Prior to our arrival, the total number of videos viewed on REDACTED.com, over the course of the previous ten months, was 370,390.  That’s essentially 9 million more views in just a quarter of the time.

She reached out to me a month ago, claiming nearly 5 million views for her new show. That’s a huge number, dwarfing almost everything else I know besides Dr. Horrible, so I naturally followed the link. I found a derivative and boring Rocketboom clone, focused on small business. The show lacked both insight and engagement. The host was blonde, which counts for something on the web. But she stayed fully clothed, further casting doubt on her show views.

Now I’m used to funny numbers. Over the past year at Revision3, we’ve learned to divide claimed show view numbers by four or five to get somewhere near the actual viewership. So how did they get five million views in a month? Time to do some forensic accounting..

So I popped by REDACTED.com, and the answer was right in front of me. In the gutter, or rightmost column, hard-up against a series of web ads, was a tiny video player. And autoplaying inside: my Joanne Colon wannabee. With a bit of clicking I verified my assumptions – that player showed up on every page. How many page views does REDACTED.COM get in a month? That’s right, about four and a half million.

Those aren’t views. Those are ad impressions. Pretty lousy ones too. I guarantee you that 99% of the visitors to REDACTED’s site came for an article or research, not to watch this breakthrough video show. But based on current IAB standards, she’s probably delivering close to the numbers she claims.

Why? Here’s a dirty little secret of web video metrics. If the video plays for more three seconds, the IAB lets you claim it as a view. Three seconds! Load up your favorite site. Count to three. Did it even finish loading? Could you have located a postage-stamp sized ad unit in the gutter before those three seconds were up? Probably not, even if your computer speakers were blaring video-noise at you.

What she’s doing is commonplace. Google has been stuffing Seth McFarlane’s unevenly funny web series into ad-units across the web. Broadband Enterprises planned on doing the same thing with the HP-funded Dorm Storm (although I never did hear how that one worked out).

Here’s another example.  I just connected with a well respected Internet video production company, much like Revision3.    His new online show was delivering more than two million views, he boasted.  I congratulated him, but secretly wondered how a very niche show could possibly deliver that many video views per month.  But a quick visit to the show’s extremely well-produced web site opened my eyes:  he was talking about "page views", not video views.  And when it comes to engagement, yield and value per view, the two couldn’t be more different.

These funny numbers are bad for everyone. Advertisers who attach themselves to shows touting this fiction will see terrible performance, and will hesitate to invest again. I know of at least one advertiser who dumped $800,000 into a year-long campaign – mostly CPA based – that failed miserably.  Here at Revision3, by contrast, similar campaigns are going gangbusters, but we’re being much more prudent in how we report.

So let’s make 2009 the year of honesty and transparency in online video metrics. It starts with the IAB and YouTube.  I’m publicly calling on those two organizations to provide real guidelines and transparency for online video metrics.  Because where those two go, the rest of the industry is sure to follow.  But until they do, everyone else will be a competitive disadvantage, unless they choose to deceive as well.  But if we want web video to continue to deliver on its promise, one thing is for sure.   The lies must stop!

Tomorrow’s post will detail why raw volume is less important for web video than traditional TV or web, and will explain a new, emerging way to value web video’s market value.

CES Smells Like Teen Comdex

I’m on my way to CES in the morning, and I’m glad I’m going.  Why?  Well apart from seeing old friends and not sleeping, it may just be the last CES ever.

News reports have been pretty bleak, with total exhibition space down from 32 to 29 football fields worth, and attendance predicted to be down as well.  But that’s not why I think CES is done – for me it’s a matter of pattern matching.

It feels like Comdex back in 2000.  I remember the show well – the biggest ever, packed with people, but little more than 3 years later it was dead.  Here are some of the parallels between the last days of Comdex and what feels like the last days of CES.

  1. Attendee Backlash: Back when I first started going in the early nineties, attending Comdex was a much coveted perk.  But just 8 years later it was a chore.  Instead of being jealous of people going to Comdex, it became common to hear, "You lucky dog", shouted at those who managed to avoid going.    Well the same thing has happened with CES.  Back in 1999 CES was a privilege.  Monday at MacWorld a group of us were talking with Cali Lewis of Geekbrief.TV, enviously calling her a lucky dog for avoiding the trip to Vegas this year.

  2. Lack of Innovation:  Comdex was a blast when computers were changing and advancing every year, but by 2002 all we had to look forward to was Windows Vista.  Nuff said.  There’s been little really new at CES over the past few years, leading to a vast wasteland of me too products and gimmicky gadgets.

  3. The Rise of Meeting Spaces:  In the early days of CES and in Comdex, all the action was on the show floor.  But just as with Comdex in 2000, you can now go to CES, see lots of new products, meet all the people you want to see, yet never go within 2 miles of the Las Vegas Convention Center.  More and more companies are opting for suites in the big hotels on the strip, and avoiding the show floor altogether. 

  4. Diminishing Cab Lines: Although it may seem like a lot of people are at CES this year, there are tons of hotel rooms, rental cars and cabs available – at least most times.  I booked a hotel room within 2 blocks of the show yesterday for $80 a night.  The MGM, Palms, Bellagio, Planet Hollywood and many other posh hotels have all been booked by friends and co-workers — within the last 3 days — for under $150 a night.  The $400 a night hotel room, for smart travellers, is a thing of the past.

  5. Extravagant Booths:  I remember back in the waning days of Comdex, booth excess grew to an outrageous level.  Graphics software company Micrografx even built an entire half-pipe in their booth and had roller skaters shredding up and down all day.  That quickly changed – by 2002 it was austere and simple.  CES has been going through the same thing. The last few years have been a cavalcade of "can you top this".  I’ll bet this year it’s all business.  The same thing happened at E3 before it died – million dollar booths that didn’t make it past the eagle-eyes of the bean counters, as the economy worsened.  Bad-ass booths are a sure sign of the apocalypse.

  6. Industry Contraction:  During the eighties and early nineties there were literally hundreds of PC makers.  By 1999 there were really only 5 that mattered.  Same goes for PC channels.  A thousand small and large stores competed to sell PCs and software, but by 1999 there were only a handful left.  The same thing happened in cableTV.  Up until 1998 or so there were a thousand cable operators, which made the Western Cable Show a drunken revel.  But with just five national pay TV companies, cable networks found it more cost effective to have five meetings around the country, rather than build a big booth to have those same meetings.  Consumer Electronics is in the same boat.  Wal*Mart, Best Buy and a few others are really all that matter.  Why build a big booth, pay for all that travel, and waste all that time when all you really need to do is travel to Minnesota and Arkansas?

Sure, there may still be a lot of exhibitors, but that’s because of the unique economics of trades shows. They are one of the biggest trailing indicators out there.  To ensure decent floor space, companies have to lock in — and pay for — their space a year in advance.  At healthy trade shows, the busiest office houses the exhibition sales staff.  And since companies have already paid for their booth space in advance, instead of pulling out altogether, they simply cut back on the booth, and cut back on staff. 

I think we’ll see a lot of simple, understaffed booths this year.  And even though we’ll never know, I’ll bet that advance sales for next year are way, way down.  Because that’s what happened with Comdex.  2000 was great, but by 2001 few companies re-upped, as the tech bubble had completely crashed all around us.

I’m not happy about it.  I still love CES, and still consider it an honor to attend.  But my pattern-matching senses are firing like crazy.

See you in Vegas – at least one last time!

Stay Out of My Ear

normal_complete_meltdown Scott Adams, author of Dilbert and a very entertaining blogger, just put up a speculative post imagining new features for your cellphone.  Most of them are very good, but one of them is way off base.

BRAIN-EXTENDER: Google and Wikipedia are already brain extenders. You can find almost any information you want and quickly. But imagine how much cooler it would be if your iPhone headset was continuously monitoring your conversations and answering your questions as they arise, or whispering suggestions in your ear. That application seems likely to me.

OK, this was obviously written by someone with little or no experience with an IFB – also known as an Interruptible Feedback device.  These are those little in-ear speakers, used by TV producers to talk to their live television show hosts.

You’ve seen them before – they’re the flesh colored inserts, with spiral connections, that occasionally show up during sports events, news shows and other television programs.  I’ve worn them many times, both when I was hosting shows, and guesting on news shows including CNN and MSNBC.

They can be very disconcerting, especially in the hands of an imbecile or sadist – and there are more of those producing television shows than you might think.

There you are, demonstrating a cool new product, or having a spirited conversation with Al Roker, and a disembodied voice suddenly cries out – inside your head – "tilt it down", or "there’s something hanging out of your nose".

At that, all but the most seasoned TV professionals will freeze up, lose focus, and start exhibiting that well known "Deer in the Headlights" expression.  It took me years to learn how to either completely ignore the voices, or give them just a tiny, tiny piece of my awareness.

Let’s face it – most of us can’t even safely talk on the cellphone and drive at the same time.  Imagine if we had a little voice in our head, 24/7, giving us hints, answering questions, or babbling insanities.  It will truly drive us all insane.  And mark my words – the State of California will quickly ban it while driving – and the entire population will ignore the ban. 

I may have learned to live with occasional voices in my head, but I guarantee you that most of you will not.  Remember, that at least until today, hearing voices in your head meant a quick trip to the funny farm.  And when we’re all a little "funny" in the head, I’m outta here.

I Can Haz OpenID?

who-am-i A couple of days ago I stopped by Fred Wilson’s great A VC blog.  I was living vicariously through his holiday vacation with his family in Europe, Paris and Milan, and was inspired to leave a comment.

Fred’s blog uses the Disqus comment system, which requires you to be verified before you can leave a comment.  I’d signed up before, but like many things I’ve forgotten my user name and password. I crafted my response, tried to post it, when confronted by the Disqus log in system.   Since I couldn’t remember my password, I opted for one of the other login options, in this case via my Google account.  Ooops, this is when the going got tough.  The Google connectivity software piece was apparently down, which caused Disqus to fail in a most unpleasant way.  The software munched up my post, and popped me back into a Disqus login screen.

But since I had no Disqus login, I was forced to recreate my post, and try to login again -  this time using my AIM ID.  Same problem.  Comment gone, and it was back to a Disqus fail. 

I’m obviously a glutton for punishment, since I went back and tried again, this time with my Yahoo account.  Again, my comment disappeared and I was pushed back to a Disqus login.

OK, I was not happy.  I’d written the same comment three times and it disappeared each time.  Although it appeared I could login via other, standard accounts, that functionality was broken.  A little digging, though, I and I discovered the problem.  Disqus was working with a company called "Clickpass", which delivered OpenID style functionality.  But Clickpass, apparently, was not working – and via a complex, Internet-version of telephone tag, I ended up writing the same comment three times, and in the end still didn’t get to post it.

It’s not as if the comment was brilliant.  It was some inane rambling about Europe and Milan – and the interweb is a better place without it.  No, what’s really got me riled up was the inexcusable programmatic behavior.

Look, even if your underlying software isn’t working correctly, what ever happened to the graceful fail?  Here’s what should have happened:  when Clickpass failed, which then caused a no-login event for Disqus, the commenting routine in Fred’s blogging platform should have retained the post, before giving me other options to log in.  What if I had, actually, written the greatest blog comment of all time?  It would have been lost for good.

And that gets me back to OpenID.  I love the idea of having one set of identification credentials that I can use around the web.  If it all works right, it’ll be awesome, birds will sing and the swallows will return to wherever they’ve disappeared from.  But it won’t all work right, not all the time.  We’re talking software here, and the internet, and the egos of childish web developers.  Occasional (or more often) fail is guaranteed.

It’s even worse than I feared.  A few days after my Disqus debacle I was talking with a developer friend of mine who was bemoaning the sorry state of OpenID implementations.  It seems that all the big sites have their own flavors, and the OpenID foundation just doesn’t have enough clout to force a single standard across the web.

That’s a bad state of affairs.  It guarantees more fail – and also guarantees epic finger-pointing.  Who will lose?  The users, first, who won’t be nearly as patient nor accommodating as I am.  But in the end the whole glorious promise of OpenID will be left in tatters, and we’ll be back to our walled-gardens of identification.  And that’s just too bad – because an open, interoperable identity system is actually one of the best ideas I’ve heard in a long time.  Too bad no one can get their act together to actually build it right.