21 July 2012

Social Seating Why I Think Its Bad.

Over on Facebook (on my wall and privately) several people commented on my stated aversion to Social Seating. I have my tweets set to appear on FB which given the number of tweets I do corresponds to a fair amount of daily traffic. Thus on FB and on Twitter I had a fair number of people see the post/rant.

Here are some examples of the comments I received:

On the positive side:  Could not agree more!

On the negative side:
(Humour)The professor will get seated next to John Candy
(Professor) Unlikely …But if I do.... then does that mean I get Darryl Hannah as my girlfriend?

And then there was Angry. I won't identify him.

(Angry) I call bullshit on your stated opinion, Professor. Are you trying to tell me, that given a choice (and please forget about the technology and labels), you do not want the chance to NOT sit next to someone that will piss you off? For example when travelling for business, I do not want to sit next to a kid, someone who has kids, and old lady, a fat guy, or anyone who wants to talk to me. If I have to interact with someone, I would prefer that they had something interesting to say ... and the only way to predict and/or influence the possibility of my preference is to tell the airline what that preference is. Another way to look at it: if you could hang around with all the other passengers in the lounge for a while, and choose who to sit next to after meeting and chatting with them, would you not want to make a choice? It's what we naturally do in life every day.

(Professor) Dear Angry

So let me call you on that. The assumption is an arrogant one to make. The situation is that its fine for some people but not for me. So I am making it clear I dont want social seating. For it to work - there must be FULL opt in (not casual) HOWEVER normally I am required to opt in as part of a larger package which is the way most opt-ins are handled). If there is a personal relationship - sure... but then I know that already and if I want to I can use other tools (such as Tripit) for this. BUT to blithely assume that I want to (choose my seat mate either explicitly or implicitly) is a bridge too far. The complexity of just the interaction of what happens if you dont like the person or if that person LIES in their profile (yes they do...) is too much of a risk.
Now shall we go even further... what about the liability of the provider?

‎(You told me to ignore the technology and the other procedural issues so I have done).

Your argument that someone will piss you off works both ways. So that argument is also completely spurious.

(Angry) So ... you would rather have no input into who you sit next to?

(Professor) Bottom line - I am ABSOLUTELY fine with having no input on WHO I sit next to. What right do I have to ask for it and what are the attendant issues of it if I did? - Getting it wrong and having to live with the consequences would actually make me want to be mad at someone - likely culprit? Oh yes the airline. The chances are that the components of the decision making will result in poor choices at best, IE results with VERY few chances of success. Let me enumerate them for you. Your analogy of choosing the person as a natural act is well BS. I fly on LCCs a lot. So I COULD potentially choose who I want to sit next to on these flights. Do I want to? Have I ever? Has anyone else wanted to do that with me?

But let me continue…
The algorithm of matching people is likely to be poor until it works on very large numbers and will also be inconsistent based on the type of trip I am taking (Think Pandora).
The requestor and the requestee would have to give enough data for it to be meaningful far beyond what is in our profiles today.
The terms of reference would need to be clearly understood by all concerned.
The opt out complications would be hard to manage and most people would give up before agreeing to it.
We have to remember that this is just ONE of many elements that go into the decision making of the particular seat and flight. - FOR example I really want to sit next to a young attractive girl who will make out at the drop of a hat AND wants a window. So that means BOTH of us want a window. Which would be MORE important - the window or the young lady?
So frankly I dont think the obnoxious person who sits next to me (and of course the people who think I am obnoxious) really wants to be sat next to. The RISK REWARD equation is not in balance at all.

Let me also go back to the technology piece which I was not allowed to use in my argument. I think that does not represent much difficulty at all. But the procedural ones do.

The point above about the window is just one. I could go on with all the other choices that I could be making. For example I really want to sit next to Steve Jobs but his personal hygiene habits and food fads would piss me off on a 12 hour flight. So how would the airline "choose" between me and the other people who wanted to sit next to late departed head of Apple?

How about the roulette challenge? Single or multiple people on the same flight wanting to connect. Well that would be hard to make happen just due to the numbers involved. Currently for example there are 830 people  signed up for SeatID. Assuming each takes 10 trips a year then the chances of them intersecting on any given flight is
3.45833E-06

Not a huge number – even if all these people decided to do this and then SeatID was able to increate the number to let's say 10 million users of whom 10% actually wanted to do this AND flew 10 times a year. The chances are still so infinitesimal  (0.4%) that it makes no sense. Oh yes that is the chance that two people would actually BE on the same flight before ANY other consideration happened.

Today there appears to be several airlines who are offering this service in some shape or form. KLM, Malaysian, Estonian. Air Baltic.  One who actually does the opposite. (Air Asia X). There are several sites that attempt to offer this.

Even if you tool ALL these sites uniquely and tried to get them to give you the number of passengers – it would still be less than .1% of all flight trips.

I fly a great deal and I get resigned to being seated in coach in SBC (Screaming Baby Class). Knowing this and all the opportunities – would I go out of my way to find a solution to choose a seat mate? NO? I am decidedly not the target market. Even if I was – I would be looking at sleep and work as the more important factors along with location and convenience. All this WAY before I was interested in a seat mate choice. http://www.imdb.com/title/tt0093748/ even John Candy is an acceptable hazard.

Cheers

19 July 2012

Dark Clouds and Hostage Data

Senior IT execs in the travel industry are starting to see dark clouds gather over  the always-on world of cloud computing, mobile apps and big data. Like Cereberus the three headed monster who guards the underworld there are three major threats that are beginning to rear their ugly heads to guard the gates to hell (or is it nirvana?):

·       Shifty Cloud Store Providers
·       Greedy Telcos
·       Data Hoarders

These threats need to be considered when moving travel technology infrastructure away from traditional service capabilities. So lets look at the situation piece by piece, assess the threats and then make smart decisions.

These Clouds don’t have Silver Linings

Part of the problem with cloud stores, which are still relatively new, are bad contracting, poor SLAs and vendors who have not done a good enough job on creating backup paths for the data and apps stored within the vendor’s cloud environment. Further, early cloud vendors have obviously been reading the GDS playbook and created contracts that make exit  or migration to another vendor very tough.

Are you hostage to your cloud computing vendor? The Washington Post's CIO Yuvinder (Yuvi) Kochar thinks you well could be and he is on a mission to get that changed. Three years ago, he spoke  with tech media site CIO Insight about the need for coherent price and license schemes for cloud services  http://www.cioinsight.com/c/a/Expert-Voices/Washington-Post-CTO-Yuvi-Kochar-Wants-Cloud-Vendors-to-Give-Him-an-Exit-Strategy-624540/?kc=CIOMINEPNL07112012  Now as the first few multi-year licenses have come up for renewal, Kochar raises concerns about ensuring open contract provisions are in place, which would help ease a migration from your current vendor to other vendors who may have new and innovative service provisions. We know that these provisions are a sore point for many a GDS user who has tried to ease the currently difficult migration from one vendor to another.

But that is not all, as many cloud users found to their dismay on the night of June 29th 2012, when a violent storm on the US east coast knocked out several large cloud providers. The cloud infrastructure, like any other hosted facility, can and will go down. But wasn't cloud computing supposed to be the end of all that risk, the panacea for users who were sick and tired of the vagaries of computing infrastructure?  So much for non-single point of failure.  http://news.cnet.com/8301-17852_3-57464342-71/modern-life-halted-as-netflix-pinterest-instagram-go-down/?utm_source=Room+77+Newsletter&utm_campaign=db60e7f0ba-Room_77_Outage_Update_6_30_20126_30_2012&utm_medium=email
Hotel newbie Room77 was one of the first to let their customers know about the problem. But this was a wake up call for all concerned that has sent users scrambling to look a their contracts and SLAs, because these contract and service provisions matter much more in a cloud environment.

Telecommunications Vendors as Robber Barons

Mobile users and BYOD (bring your own device) has generated a big swing to mobile (and by that I mean non fixed location) services. Just as everyone gets behind this – the users, the equipment manufacturers and the applications are all taking advantage of an unlimited wireless world. BUT there is that small matter of mobile data plans.

Consumers and business users alike are being besieged by greedy telcos who are  charging exorbitant rates for data roaming plans.  (Albeit this is subtle as most users have not seen the changes occur. The mobile providers change their contracts with such frequency that the users are truly confused and in the end most of them have not cared). This is caused by the twin problems of poor understanding by the user of their data plan pricing and very high levels of expectation set by – yes you guessed it – mobile telecommunications vendors. No TV commercial these days worth its salt will exclude the promise of steaming video to your mobile! How many of you have been stung by the problems of:
·       Slow mobile speeds
·       Blank data spots
·       High data plans costs
·       Usurious rates for international roaming
My point is clear – these are the real world and the promise from the mobile operators is far from realistic and of it can get course VERY expensive.

It's Mine (Data) and You Can't Have It!

The value of mining big data is seeing everyone  who is a data processor (and his mother) trying to get protective about the large volumes of data that passes through their infrastructure. In turn these data processors now think of their custodial duty no longer as just a cost but also as a money spinner for them

Data was once considered to be so difficult to mine that holders of data didn't mind sharing. Now that there are affordable and effective tools for manipulating big data, all of a sudden those controllers of data (note not the owners) are being really protective about access to  this data. They are hiding behind "privacy" and "security" concerns when what they really mean is that they want to get the commercial value from the data, and not give it to the ones who are smart enough to think of ways of using it. I don't want to belittle the issue of privacy and security concerns for personal data but there are very easy ways of managing those concerns.

As an example look at large scale data processors today who make considerable sums from the data they are processing. Imagine the chagrin of the GDS vendors who are about to see their current cash cows of MIDT contracts (the sale of travel agency behavior data back to the airlines) go out the window to newer smaller companies now able for far less cost can both capture and create useful information from this raw data.

So be warned out there. It looks like there is a new set of guardians on duty as a Gatekeeper. And we had all better take care. Be prepared to make the journey through the underworld to the blissful state of contentedness that the promises of Big Data and Cloud Computing should be taking us to. And that nasty looking puppy just might prevent you from getting there.

Cheers