Amy Wohl's (Archival) Opinions on Cloud Computing and SaaS

Blog Announcement: I'm Consolidating

For nearly 20 years, I’ve been blogging at two sites: Amy Wohl’s Opinions and Amy Wohl’s Opinions on SaaS and Cloud Computing.  I’m consolidating all my blogging onto the other blog, Amy Wohl’s Opinions.  I’ll continue to blog on Cloud Computing and SaaS there, as well as all my interests related to bringing new technology to market.  I’ll be actively using the blog categories there to help you find specific topics, including (but not limited to) Cloud Computing and SaaS.

I've added a new feature at my other blog, Amy Wohl’s Opinions, too, so that you'll be able to subscribe and get an email each time I post.  You'll find that on the RH column towards the top, at Amy Wohl’s Opinions.  

I hope to see you soon at Amy Wohl’s Opinions.

Note: This blog will become an archival site only now.

More on HP's Reinvention

Comments and analysis from all over the industry continue to pour in.  The general opinion is that HP is trying to correct for the mistakes it made in past years and that their track record says they'll just be making new mistakes.  A consensus seems to be forming around

(1) Management mistakes like trying to follow trends but coming in too late with the wrong skills

(2) Lack of focus -- trying to do everything; even big companies can't do that

One very good article I found hones in on the mistakes HP made and how partners can identify and avoid such errors.

What is HP Thinking?

Officially I'm on vacation, but this seems like a summer without vacations!

I thought of heading this post with the title "Is HP Crazy?" but I decided that didn't quite make my point.

I've just read my way through a dozen posts (try this one from Dennis Howlett) on what HP just did -- announce that it was closing down its nascent WebOS (tablet and phone) business, say that it was evaluating whether to stay in the PC business or spin it off, and announce that it was buying British enterprise software company Autonomy for an amazing $10.3 billion.

I think the press may be covering these in the wrong order.  Here's another way to look at it.

HP must have been working on the Autonomy deal for some time.  It wanted to announce it now and use the idea of "we're becoming an enterprise company" as the cover for everything else. 

That would be nice if it were believable.  HP has been trying for some time to move into the arena of enterprise software and services, but they have yet to be a big enough player to gain critical mass and buying another niche software vendor  is unlikely to give them the traction they're looking for.  Of course, they may have other acquisitions up their sleeve.  I am concerned that they are paying too much.  Also, deciding to keep Autonomy separate which will guarantee that they can't sell integration as a benefit.

HP then  looked at their business and must have thought one of the reasons we're not taken seriously enough as an enterprise play is that too much of our business is PC oriented.  If we take that business off the table, our enterprise business will immediately become a much more important part of HP.  I think someone should have been asking how they were going to make up for the revenue loss on their PC business, but that doesn't seem to have been in the mix.

In the future, when we understand the tablet business better, we'll be in a position to judge whether their decision to bail out without ever having given the WebOS-based products a real chance was genius or a really bad idea.  So much of the tablet business right now (before the enterprises start buying tablets in volume for mainstream applications, HP) is all about competing with Apple -- and HP is in the business of building products but not so much being a consumer marketeer of premium-priced products.  I thought that the tablet business was a good fit for HP's drive to the enterprise, but apparently HP did not see it that way.  Perhaps they should have been less ambitious and positioned their tablet product right at the enterprise rather than at consumer markets?

I am putting their decision to announce that they are going to spend the year or more considering whether or not to sell off their PC unit under the "doing an Osborne" category (preannouncing something and freezing your current market).  Why would anyone buy an HP PC (or do a partnership deal) when this black cloud is hanging over the HP PCs future?  The ultimate value of HP's PC division will be plummeting while HP considers its options.  If this were the point of the recent announcements, it would have been accompanied by a statement that HP was in talks with XXX and that the details of the deal would be announcement shortly.

So I'm disappointed with HP.  Not with the fact that they're probably years away from becoming one of the top enterprise system companies.  Not with their initial poor results in the tablet business.  Not with their desire to be out of the low margin PC business and into higher margin businesses like services and software.  Rather, I am disappointed with the way they are going about it.

As it is, HP is going to provide cheer for tablet and PC vendors who now have a bigger market to divide among themselves and continue to fuel the already hot market for enterprise software, a market where there are already bigger and more successful players.

A Speculation on Barriers to Cloud Computing

My good friend and colleague Michael Salsburg of Unisys sent me a new blog of his and I can't resist sharing it with you.  I can't link to it because it's not on a publically available site, so I'm just going to include it here. 

It's mainly about the fact that moving to the cloud is less about technology and more about human barriers.  I couldn't agree more.

Comments will be appreciated.

Stop Dave, I'm Afraid

Over the last week, Tom Mallon and I have visited a number of customers and prospects in the Asia Pacific region to discuss Unisys’ Secure Private Cloud.  It has been both exhilirating and instructional on both sides of these discussions.  We have engaged IT directors and CIOs in interesting discussions and we have been quite proud to discuss the depth of our SPC solution as well as the experiences we have accumulated over the last three years in creating and managing a cloudy environment.  We are definitely providing thought leadership and bringing up ideas and suggestions that go far beyond what the customers and prospects have considered.

But trasforming portions of a datacenter into a cloudy environment is way more than technical in nature.  As a matter of fact, the biggest hurdle has appeared to not be of a technical nature at all.  Each CIO or IT manager expressed a very similar hurdle that they are facing.  It’s essentially their stakeholder’s fear of automation and reluctance to lose control.  Thus the title of this blog.  Perhaps many of you remember this blog’s title as a portion of dialog in“2001 – A Space Odyssey”.  I can’t say I understand all of the messages in the movie, but the theme of automation gone wrong is quite clear.  Only when you hear HAL say “Stop Dave, I’m Afraid” do you realize that the fear of loss of control goes both ways.

A number of our SPC conversations touched on a concern about creating and managing the huge number of VM images for their developers because each developer would request their own customized VM.  Developers were not comfortable with the standardization that goes hand-in-hand with automation.  We explained that, when we first started automating in the ERL, we were in a simular situation.  Provisioning was considered a custom request and required a minimum elapsed time of two weeks to be satisfied. Today, 95% of requests in the ERL are “standard” and, through automation,  makes a VM available within 5 minutes.

This gave them courage.  Other discussions were more general than addressing the customization issue.  An IT director explained that he was calling their cloud initiative a “cloud architecture” since their stakeholders were very uncomfortable with centralizing control of their computing into a cloud that is shared by multiple business units.  And so it went throughout the week.

I was reminded of a situation I found myself in a long time ago.  I was architecting an automated admissions system for a college.  The admissions officer had a box on her desk that was filled with 3x5 cards.  She literally but her arms around the box and said that she could not possible work without it. 

So, as we worry about which cloud technology is best for our customers and fixate on vendors that compete with SPC, we need to understand something that continues to be the maxim regarding datacenter transformation.  The real competition to our solution is not technical.  It’s inertia and an inability to articulate the value of overcoming this inertia in the eyes of the stakeholders.. 

Of course, automation is the inexorable trend in our industry.

Nevertheless, customers will not respond to the attitude that “resistance is futile.”

Michael Salsburg

 

A Backbone for a Social Netwrk

It's much too soon to know what the effect of Google+ will be on Facebook or other social networks, but I'm betting there are lots of people who may transfer their social conversations to Google+ if it is and becomes as open as it seems to be.

There's an excellent article on the O'Reilly site, explaining Google+ and speculating on how it might be used.  I recommend it.

The issue is that a social backbone, supporting many kinds of social interaction, must be open.  Users must control who gets what and be able to take their data elsewhere if they choose.  Facebook doesn't work that way -- but Google has no need to be a walled garden.

I can't wait to see what happens next.

 

Is it Illegal to be very successful?

Watching the FTC start its investigation of Google does remind me of earlier investigations of success (such as the Microsoft Explorer browser).

As I understand the law, there is nothing that prohibits a company being so successful that it has an effective monopoly and controls a significant portion of its marketplace as long as that company is doing nothing illegal in the process -- especially not preventing other companies from entering the market and customers from using competing products.

I don't think Google does either of those things except that their success feeds on itself -- the more successful they are in their search business (which is what this is all about), the harder it is for others to be successful search companies (although Microsoft's Bing is having a fair try) and the less likely users are to try a different search engine.

Note that it is not quite like the auction issue -- where the larger an on-line auction company becomes the less attractive it is to list your items on another site because fewer people will ever see it.

For niche things (fine art for auction sites), specialized topics for search engines (medical, for example), there may, in fact, be better places to go.

Google is interesting because (1) it's so big and (2) its success in search, which allows it to collect enormous revenues in ads to the searchers, funds all of the free software which attracts users to its site -- not just to search but for everyday work.

I cn't imagine an Internet without Google (although this may merely mean I'm lacking in imagination) since I have not seen a better way to find the things I seek quickly.  But someone else can try and in the world of the Internet it would be possible for users to find them, in volume, quickly.

What will happen -- too soon to tell.  But we will all be watching.

 

Do You Have a Book Reader Yet?

As a devotee of book readers I’m now on my fifth Kindle (two broken and replaced by Amazon, two lost and replaced by me).  I’m at the point where I’m unhappy if my Kindle is not in my purse or tote bag.  But then I’m a dedicated reader.  Before the Kindle, I always carried at least one book with me.

To me the Kindle is a kind of miracle.  It lets me carry many (3500 in the current model) books with me so I can follow my usual practice of reading two or three books at the same time – something serious, something literary, and something light (a mystery or some “pop” thing).   Of course, it’s most useful when I travel but it’s also useful at home, to avoid being annoyed with wait time at doctors’ offices or in supermarket lines.  As I said, I’m a reader.  I learned how at three, and I’ve been going ever since.

(Historical Note:  I tried out many book readers before the Kindle, as they came along.  None of them were useful enough for the inconvenience of lugging them along, keeping them charged, or dealing with their limited libraries.)

Recently, Esther Schindler, a freelance writer, editor, and analyst, noted that she wanted to know about what eReaders would look like in the future and how you would tell it was time to buy one – or time to buy a new one.  That seemed to me to be a good question to think about.  Here are my thoughts:

On Book Readers Generally

• Book readers have come down considerably in price.  You can know buy an Amazon Kindle (not the model I use) for as little as $114.  If you’re a book every few days reader like me, you could probably save the cost of the reader in the discounted price of the books you’d buy anyway in a month or two.
• I’ve already bought into the convenience of book readers, but not everyone has.  I still get people coming up to me in public and asking me about my Kindle.  They often say “I can’t imagine reading books on some kind of electronic device.”  Actually, in about 30-60 minutes of usage, the book reader disappears and you’re just reading the content.  There are some neat features you might choose to use (bookmarks and annotations), but I generally just read.
• If you’re a member of the aging population, you might enjoy the fact that book readers, being digital devices, can let you adjust the type size to your eyes.  It certainly makes me happier.

 On Future Book Readers

• Color will be available on most book readers soon (it is already available on some, like one version of the Barnes & Noble Nook).  If you’re used to reading ordinary books you may not really miss color, but it’s required for most children’s books and books with illustrations (medical and science textbooks for example).  Color might limit the usage time – which is a big consideration for me.  When I’m on the road I can’t always recharge my reader every day, so the fact that it will happily run for much longer is important to me.  It will also probably cost more – whether color is worth it will be up to individual users.

• Multi-function devices:  Many have said they’d like a reader but one that they could also use for other things, especially Internet access.  Some readers have at least limited Internet access already; they could have more.  If you want to read books on a multi-purpose device you can choose to use a SmartPhone or a tablet.  When I lost my Kindle on a trip to California (we had to change planes because of a mechanical failure and I left it behind – the airline couldn’t “find” it) I just switched to my Android phone.  Now I would probably prefer my Android Xoom tablet.  Both of them have Kindle readers and can synch with my Kindle (wherever it happens to be), offering me the book I am reading on just the right page. 

A note on reading on other devices:  The reason I like the Kindle is that it’s small and thin and easier to hold than a paperback book.  My Motorola Droid X Android phone is small, too, but it’s much harder to read, with its smaller screen.  My Motorola Xoom Tablet has a beautiful screen, but I wouldn’t want to hold it in my hand(s) for hours on end.  For now, I’m opting for separate devices – I do think this will change.

By the way, an important use for ereaders is text books.  They’re really expensive and they’re big – carrying a stack of them around is an invitation to a back ache.  But, just like the sharing problem noted below, the economics of text books (people resell them and buy used ones) will have to be figured out.

Further Futures

Go out five or ten years and we may be able to have some really interesting choices.  I’m thinking of a foldable reader, like a sheet of paper, that I could unfold and use anywhere.  The original screen technology from Xerox PARC and from E-Ink could allow for this kind of futuristic “hanky” configuration.  Usually, it’s discussed in terms of a self-updating personalized newspaper but I would think it could be designed as a bookreader.  Just think!  It wouldn't weigh much of anything and its size would allow for a "real" book page and even some annotations or drawings.

IBM and others have done research on technology that projects the contents of a computer screen onto special glasses.  There’s no reason why this couldn’t be the book you’re reading.  They’ve also looked at a little lens device that you wear (perhaps attached to your glasses) that projects a screen into a virtual screen of any size – so you could read your book on a 6 foot display, in really big type.  This might be particularly interesting for illustrated books.

There are two things I don’t like about bookreaders (other than the fact that I’ve broken or lost four of them so far).

They don’t really let you share books.  In the past I gave away about 300 books a year, as I finished them.  There are ways to share ebooks but they are clumsy and limited.  It should be possible to solve this problem.  If I shared books I the past, why can’t I pass them on now.  It should be the same rules – I’m passing on my access and I can’t get the book any more.

There are books I still buy in the physical world for different reasons:

Collections:  I have a collection of science fiction books and one of cook books.  I don’t want some of my books to be in the ereader and some of them on the shelves where I can look at them in groups, for example.  I don’t know what the answer is to this one.

Art Books, Pop-Ups (I have big collections of each) don’t fit into an ereader and I’m not sure they ever will.  So they continue to take up lots of space.

I think ereaders have gone from being geek toys (I used to be able to guarantee meeting anyone near by if I took one out and turned it on) to being mainstream.  If you read a lot, you probably should have one.  If you read and travel, you definitely should consider one.  I have never bought the newest model when I still had one I was using -- but I have had involuntary upgrades to keep me up to date.

Will the NFC Wallet be the Winner for Custom Information and Mobile Payments?

Google has put its considerable weight behind the concept of an NFC (Near Field Communication) wallet.  (That means a system that lets you use your SmartPhone to receive customized deals and pay for merchandise by tapping your device to a special reader.)  Everyone I’ve interviewed stresses that while it is a payment system, that’s not the point.  The real importance is that it will know “everything” about each user and provide opt-in information about interesting merchandise and discounts based on the user’s location and preferences.  Users will also be able to tap their “wallet” (an enabled SmartPhone) on a sign as they enter a store or mall, and see special deals available now.

It’s an interesting idea based on the facts that:

• Nearly everyone that a merchant would want to connect with is walking around with a mobile phone, and phone manufacturers are expected to include an NFC chip in their new smart phones.  Some have already started and it’s expected to be nearly universal within five years.
• Many large merchants have already agreed to participate (which means putting NFC chip readers throughout their stores and creating personalized deals – Google intends to help with that).  Google has signed up 16 chains, including Bloomingdale’s Macy’s and Radio Shack.  Of course, they have to be enabled with the SingleTap technology and put together their plans for offering “deals.”
• MasterCard (but not Visa, which has a competing project under way) has agreed to be the facilitating processor for verification and payments.  CitiBank and First Data Corp. will also be involved.  A carrier is not required for the scheme to work, but Sprint is part of the Google initiative.

Google has emphasized that the initiative is open to anyone – that means, of course, many merchants, but also more banks and carriers and more SmartPhone platforms than Google’s Androids.  Google has not commented on enabling BlackBerry or iPhone, but that could be in the works.  One source we spoke to expects Apple to have an initiative in this category, but possibly a separate one.

This is far from a sure thing.  First of all, this is a complicated idea and everything has to be in place at the same time:

1. Consumers have to be educated and participate by signing up, providing their information, and looking at the deals and sometimes opting in.  Merchants will want to know what that participation rate will be.
2. Merchants in numbers that support continuing consumer interest must sign up (they need to do this is numbers because different customers have different shopping patterns.  I might want deals from Neiman Marcus, Saks, and Target, while someone else might be interested in Target, Walmart, and Shop-Rite.
3.  The system need to support a payment agent customers already use and customers must perceive it as secure and easy.  In the test phase of the Google NFC Wallet, which is just getting started now, customers will pay via their MasterCard account.
4. It needs to eventually work across all the major SmartPhones – merchants are not likely to be enthusiastic about supporting multiple systems for different SmartPhone platforms and it would be confusing to be out and about with a friend or partner and get different offerings based on which SmartPhone you happen to use.

We’ve had earlier experiments with some kind of wave or tap and go payment system.  In Europe and Japan these systems are popular and well-established.  So far, in the U.S., none of them have caught on.   This may be the problem of getting everything together in one time and place or it may be that in the U.S. there are so many options for getting customized deals (Groupon, Living Social, etc.) that it’s hard to get customers’ attention.

Or this may be the magical moment when everything comes together and we change the way we shop.

One expert we interviewed, Mick Mullagh of VIVOTech (which makes NFC readers, in addition to being involved in the whole initiative) points out that as with all new technology, this is a rollout over time.  He expects to see 20 million phones enabled in the next 18-24 months.  There are already 600,000 NFC readers out there.  You may have come across one in a New York City taxi cab or at CVS.  Merchants, different geographies, and customers all have to come to the party.  Mullagh is pretty sure that the value proposition for both merchants and customers is the customized offer.  As I said to him, “I’ll be looking to try it out” – both on the road in New York, San Francisco, and Portland, where it’s starting, and in the suburbs of Philadelphia when it gets here.

A New Intelligence to Apply

Watching Watson, IBM's semantically adept (and seemingly endlessly informed) software answer questions on Jeopardy is great fun.  We can even relate to Watson as a persona, feeling sorry for him when he gets an answer wrong (not very often) and feeling pleased for him when he wins.

But last week I got to see a demonstration of Watson doing something genuinely useful -- Jeopardy for Watson is a fun way of demonstrating some of his skills, but his real purpose lies elsewhere.

A medical doctor told about feeding Watson all the information that a medical student, ending his second year, would know.  (That's the part where you read these huge textbooks, memorize a lot, and pass tests.)  The next thing is to feed him clinical data bases (which might include what drugs are used for what, patient symptoms and records, and what causes a successful outcome).  That's a work in progress.

Already Watson can answer some medical questions.  He's not going to replace doctors, but he will provide enormous amounts of quickly processed information so the doctor can make a more educated decision. 

Of course Watson's combination of machine intelligence, fast processing, and subtle semantic processing (he can learn jargon, slang, and new vocabularies) can be applied to many fields.

Think of anything where there's a huge and growing body of knowledge and an understanding of what the rules for identifying, collecting, processing, and inferring about that knowledge might be.  Various research fields from oil exploration to pharmaceuticals (new drug gesting), come to mind.

The application I would really like to use Watson (or one of his offspring) for is the one that is the "real" version of Neal Stephenson's The Diamond Age," where A Young Lady's Primer contains all of the knowledge of the world and can answer any question.  In that case, the Primer couldn't -- a talented actress was connected to the Primer to provide the answers.  Watson today is programmed to focus his inteligence on specific domains of knowledge.  But just what he does in his Jeopardy persona could answer most of the questions a bright ten year old might ask -- when we can make him -- or his use through a personal knowledge cloud -- much cheaper. 

Personally I can't wait!

Cloud Computing is not Black Magic

Amazon's recent outage has brought up many questions about the "safety" of cloud computing.  The problem (beside Amazon's lengthy outage) is mainly that there is a lot of magical thinking about cloud computing.  It is not a conjuring trick that allows the user's organization to depend on the cloud vendor for everything, but rather a computing platform that requires thoughtful architecture, implementation, and management.

Not every public cloud is the same -- some offer mainly raw compute power (and perhaps storage and tools) -- others offer a managed computing experience.  Users need to know what they are buying and plan accordingly.  As Jonathan Spira points out in his article on the Amazon outage, companies who planned for trouble (and spread their data and computing over multiple sites) did not have a problem.

Cloud Compouting is a great way to avoid the cost of implementing and managing infrastructure; what you run on that infrastructure and how successful the results will be depends on the skill and effort your team put into the work.

Connecting to the Cloud - The HP TouchPad

HP has finally announced his webOS-based tablet, the TouchPad.  You can read a description here, and view some pictures.  But Tony Wolverton's fine article says it may be too late for the TouchPad, which won't ship until summer.

I beg to differ.

New markets can take a long time to form.  So far the tablet/slate market is mainly Apple, and a bunch of vendors (especially Motorola with its Xoom and RIM with its Playbook) who have announced products for late spring.  Apple has had a magnificent start in this market and is likely to continue with a Version 2 of the iPad coming, perhaps in the 2nd quarter.

But there's plenty of room in the market, particularly in a market we don't understand very well.  Let's keep a few critical points in mind.

1. We don't understand the tablet market very well yet.  Are tablets a replacement for laptops and netbooks or simply an alternative format for some users? So far, tablets are a small fraction of the overall portable PC market.
2. Who are tablets intended for?  Mainly consumers?  Businessmen as consumers (individual buyers)? The corporate market? 
3. How will tablets reach their market?  As individual purchases with downloaded software?  Or as corporate buys, with specially developed software for business applications?  Probably some of both, we'd guess.  This is important, because the TouchPad, with its unique OS, will need its own developer ecosystem and applications and this is more important for the consumer market.
4. How does a tablet relate to a smartphone?  Is it a replacement?  If it doesn't replace the phone function of the smartphone, does it mean users will eventually move toward a tablet plus a very small, mainly voice function, mobile phone?
5. What about in emerging markets, where the smartphone is emerging as the main connection to the Internet?

To me, the tablet is all about a world in which we need a lightweight (physically) device which easily connects to the Cloud and provides a great user experience (interface, battery life, choice of applications, etc.).  It also needs to provide an appropriate computing environment for its target users.  That means some of us need keyboards that aren't on glass and all of us need easym, cheap connectivity for this to work.

A lot depends, of course, on pricing and features.  So the new HP TouchPad seems to have new features, but we don't know much yet about pricing, telco partners, and connection costs.

In fact, we're going to have a chaotic year in the portable device market as vendors introduce at least a dozen products, some consumer, some business and some (as Apple hopes) cross-over products for both markets.  Until we've had a chance to see more of the players and offerings I think it's a little soon to look at anything that seems to be attractive and useful, backed by an existing vendor with a strong distribution scheme in place, and write it off. 

I'm taking a wait and see attitude.