latoga labs

Alliances & Partnership Advising

© 2006–2025 · Log in

Notes from Camping in the Clouds

Leave a Comment

Last night I attended CloudCamp Silicon Valley. Having recently attended VMworld, Orcale OpenWorld, and CloudCamp, I have to say that I still think there is great value in the Unconference Format.  While VMworld was one of the most productive trade shows that I have ever been at, one can only hope that it can maintain that productivity as it grows.  Oracle OpenWorld has grown into the typical trade show venue where attendees are locked in sessions most of the time and vendors sit around their booths checking email.  While the unconferences are held for the benefit of the attendees and the attendees decide what gets discussed and vote with the feet (walk to the discussions you like and feel free to leave if you don’t find it valuable).  I was a bit concerned about the event being held at Suns EBC, but it worked well overall, nice facilities for the group and infrastructure for the sponsors to show their stuff before the event began (thought there were a few predatory vendors in the group who don’t understand the the point of an unconference…I’ll admit it felt good to shoot a few of them down; I only hope they learned their lesson)

It was great running into not only old colleagues and industry contacts, but an alumni of my university as well.  Fun stuff reminiscing about old times back at UWEC CS department and the hindsight value of that program.

One of the sessions I attended was lead by Mike Pittaro from SnapLogic (a company I did some consulting work for earlier this year).  His discussion was well attended by over 15 poeple to discuss Connecting Clouds.  Essentially talking about the integration needs of cloud computing, it’s similarity to SOA, and the cyclical nature of integration along with computer trends in general.  I found how personal corporate connections came up during the discussion: my ex-collegue from IONA Steve Vinoski was referenced with his REST vs SOAP discussions; the demise of SOA vendors referencing IONA’s recent absorption into Progress Software; and the proprietary versus open standards discussion referencing VMware’s ESX hyperadvisor.  We also discussed open standards for connecting clouds versus proprietary connections between clouds, will proprietary connections emerge as the cloud coellesces from it’s post big bang state into gallexies and solar systems?  This discussion was very timely considering GNU Founder Richard Stallman’s recent comments on cloud computing.

Even more interesting the the last session I joined which was lead by James Urquhart and discussed the current legal situation around cloud computing.  Unfortunately, this was a very thinly attended discussion based on the importance of the discussion to anyone doing anything in cloud related.  We discussed topics ranging from the Stored Communications Act, legal ramifications of non-US customers to US cloud companies based upon the physical location of their cloud service, and even the concept that cloud computing and virtualization enable of Follow the Law computing (similar to Follow the Sun or Follow the Moon computing, moving your computing to physical locations that are legally favorable to the work that you are doing). There are definitely some legal aspects of Cloud Computing that all cloud companies and cloud users need to realize and get changed or risk Cloud Computing progress disappear into the ether. Very interesting night of discussions…

One of the most interesting aspects I took away from the night is the split between the Web 2.0 Cloud and the Enerterprise Cloud.  The Web 2.0 Cloud is the all the classic cloud services that you think of when you say Cloud Computing:  Gmail, SalesForce.com, Joyent, Amazon Services, etc.   The Enterprise Cloud is the continual migration of internet technologies into the internal enterprise data center along with the evoloution of virtualization in the enterprise and the move of large enterprise IT organizations into true internal service provider role.  VMware’s vCloud initiative and technology vision will enable an Enterprise Cloud to connet with the Web 2.0 Cloud and move computing from inside the enterprise out or oustide the enterprise in. But all the discussions were based around using or building around the Web 2.0 Cloud or a few vendors who were talking about the cloud computing technologies for the enterprise.  Companies and individuals need to be clear on which of the two sides of the cloud they live on.  Nothing is worse than talking on a topic for 15 minutes only to realize that each party is living on the oposite sides of the cloud.  I assuem this disparity is part of the reason that VMware didn’t official sponsor or get involved in CloudCamp; which is a bit surprising considering the new recent announcements and focus.

Screen Scraping Lunacy & Need for Open Data

Leave a Comment

Today I was talking with a fellow entrepreneur who told me about a new class of web2.0 companies that I had not heard of before.  Financial portal sites that you give your financial account login info to and then they screen scrape those sites to provide you a single portal view of all your information and then anonymously compare it to the community.

Are You $#@& Kidding Me?  (Pardon my censored obscenity…)

That was my initial reaction when I heard this.  Giving some 3rd party website my login to my social network or photo sharing site for data aggregation purposes is one thing (and that is scary enough), but who in their right mind would give some 3rd party website the login to their financial accounts?

And yet, that is exactly what Cake Financial, Covester (recently raised $6.5M in funding), and others are doing. Granted, I don’t know a lot about these firms at this point, I know Cake claims to integrate all your finanical data and Covester might be more of a money management social network (wasn’t this already tried with the messages on Yahoo Financials?  I know insiders who leaked data and caused possible financial damage to their companies on sites like that).   I am both surprised and not by the fact that these companies exist and are getting funding.  I just hope that they are thinking of the long term and learning from the recent financial scandels and are putting real safeguards in place to protect themselves, their investors, and most importantly their communities.

Time will tell…

But I do think this just shows the real need for Open Data sharing access that is the corner stone to the new value proposition for most of the internet based services.  This is an area that I have been doing a lot of thought and research on lately.  First by building a proof of concept demo for SnapLogic that integrates an open with a not-so-open service to show the added value of the combination.  And Second, leveraging that initial POC and my years of selling experience to come up with a number of interesting ways to leverage these two data types to provide disruptive changes in the sales process and possibly organization of sales teams (more on that later…).  I have also been working through a number of real business differentiating photography sharing capabilities that use the same concept for FocalPower.

The need for services to enable their uses to access and share their data and metadata with other services/users is real and valuable.  The real trick is choosing the right level of openness  to make it usable by the masses and proliferate while enabling new revenue generation.  You have to balance the openness with the protection of your users.  How do you do that such that you don’t prevent the innovation that this type of openness allows?  Look at all the innovative things that have come out of the open API on Flickr.  Opening up your service via an API to share data and metadata has to be an all or nothing proposition with regards to who has access to that data.  Placing too many restrictions just stifles the innovation.  Note I said restrictions, not safe guards.  There needs to be safe guards in place to protect the users.  And the ability to users to opt-in to the open access or different levels of the access.

The key thing in all these integrations is for the users to read the terms of service!   You may be invalidating your terms for using a service if you give your login information to a 3rd party.  And, like the ownership rights grab that have been making waves in the photography industry, you might not own all the information that you expect to own if you use a service.

Edge Integration with SnapLogic

2 Comments

Around the beginning of this month I started doing some consulting work with SnapLogic. SnapLogic was founded by Informatica alumni who were solving problems of quickly gaining access to data located inside the enterprise in order to build hybrid applications that solved their real-world problems. They originally solved this problem with resourceful scripts that would pull data out of various data sources, store and/or index this data, and then make it available to users in a format of easily consumable by the end users. (when they were originally doing this, the term Mashup wasn’t coined yet, and still isn’t favored inside corporate walls…hence the term Enterprise 2.0).

They quickly realized that they were working for a data integration firm and had build their own low level RESTful data integration. Hence, SnapLogic was formed. As an open source company, the SnapLogic edge integration platform can take data from various data sources and represent a RESTful interface to that data. Since the platform has Components that can talk to standard data sources, you can use the platform to implement a standard data services layer within an organization. Also, since it’s all open source, you can build your own Components, or use ones built by the community, to provide data services into a mix of data sources: local, remote, accessed directly, accessed via an API, etc.

On the internet, we have all see mashups of data from multiple sources (i.e., fetching a set of locations from a real estate database and then laying those homes for sale on top of Google Maps). This is an example of the RESTful data integration in action. Slowly, the enterprise has been pulled in this direction by their users (usually with the IT department kicking and screaming because they have to give up control). The SnapLogic platform has the potential to offer the IT organization a layer of control, but also offer the users access to the data that they need in order to build enterprise mashups.

I have been aware of SnapLogic since before they came out of stealth due to the fact that a number of people that I used to work with were involved with the company. At that time, I was working at IONA and there were a number of interesting discussions going around about REST (most were usually started or ended by Steve Vinoski 🙂 ) as well as the process of migrating the company from a pure closed source software provider to a hybrid model of closed and open source. Thus, I was glad when the opportunity arose to help out SnapLogic recently and gain more experience with their platform; mixing my integration experience with my recent experiences in open source.

Stay tuned for more about my experiences working with the SnapLogic and their Edge Integration Platform