Database as a Service – from Force.com to Amazon RDS and SQL Azure

These days, those who are building internet applications have many choices for database.  Traditionally (i.e. up to the last few years), you have the usual commercial and open source solutions such as Oracle, Microsoft SQL Server, MySQL, PostgresSQL to store user and application data.  Usually, these databases run in the company’s own data centers or a co-lo facility managed by AT&T, Verizon or the like.  However, offerings from Force.com (from Salesforce.com), Amazon Web Services and Windows Azure are changing this landscape.  In fact, in some cases, they start to converge despite their differences.

I started thinking about database as a service not because I was building one or looking for a DBaaS solution.  In fact, I was building a full stack internet application based on open source technology.  However, the offering from Force.com was rather intriguing as they claim to offer faster development time, easy to use language and secured infrastructure as this InformationWeek article indicates.

However, when you dig a little deeper, you will find that if you do not want your UI to look like the Salesforce.com form based interface and instead you want a rich interactive user experience, Force.com offers limited options.  The reason is because Force.com only offers a database driven form application UI or a heavy client written in Adobe AIR and Flex.  The former is reminiscent of web application from the late 1990s and the latter requires significant processing power on the client desktop.

Now you may say Force.com supports Java, Ruby and every other language under the sun.  That is indeed true though it is also where Force.com veers from a Platform as a Service provider into a Database as a Service provider.   If you want to use Java, Ruby, PHP or another language to build your application logic and client interface, you essentially have to host that code yourself.  You would be using the Force.com infrastructure for database and maybe stored procedures.  Sure you can host your application code at AWS or even Azure cloud but does it really make sense to have multiple vendors hosting different tiers of your technology stack?  It is tough enough to satisfy your SLA for applications that you run fully in-house.

In short, if you see a need to contract with multiple hosting vendors for the database, application and client technologies, you may well start looking at Force.com, Amazon RDS and SQL Azure as similar solutions.  Each has its strengths and drawbacks.  Here is an article comparing RDS and SQL Azure that you can start with.

Add to FacebookAdd to DiggAdd to Del.icio.usAdd to StumbleuponAdd to RedditAdd to BlinklistAdd to TwitterAdd to TechnoratiAdd to Yahoo BuzzAdd to Newsvine

Quick Tests to pick EHR by

This is a short article crediting another author for creating the criteria with which to select an EHR. There are so many vendors and so many sales people out there telling you their wares are the best but little assistance for the medical practitioners who need to choose these tools in the time that they do not have.

For those as confused as I was a couple of weeks ago about these jargons, an Electronic Health Record system (EHR) is a superset of Electronic Medical Record (EMR) and Personal Health Record (PHR) systems.  The stimulus money from the Obama administration to encourage the adoption of software technology in medical practices is for EHR as it requires the introduction of both clinic and patient facing features.

This is the original posting by Peter Beck from which the summary below was created:

  1. Any Information To Be Filed Must Have Doctor Approval Or Awareness
  2. External Resources To Take Care Of Scut In A Cheap, Dedicated Fashion = PASS
  3. Get Doctor Signoff Before Proceeding With System Changes — Repeatedly
  4. 1-Click Is Best, 2-Clicks Is Status Quo, Anything Else…Not So Good
  5. Pare With Care
  6. Always Customize By Provider, If You Have The Option
  7. If Something Works, Copy It Shamelessly

Add to FacebookAdd to DiggAdd to Del.icio.usAdd to StumbleuponAdd to RedditAdd to BlinklistAdd to TwitterAdd to TechnoratiAdd to Yahoo BuzzAdd to Newsvine

Why do doctors hate EMRs?

While trolling the web, I found this great article written by a physician about why doctors hate EMRs.  Electronic Medical Records, or its cousin EHR and even PHR, are supposed to help physicians provide better quality care to patients, ideally at lower costs to the society.  However, as you can see in this article, http://www.dillingerkovach.com/accusourcecareers/?p=10501, the adoption rate is abysmal.

As a software guy, and having studied cases of why Business Intelligence and other other enterprise software systems suffered poor adoption among their intended users, I can almost predict the usual barriers to adoption.  What makes the EMR case more interesting is that in small clinics, the physicians tend to be the buyers of the software too.  If they are unhappy with the product, they would simply not buy an EMR or abandon it after paying for it.   Those who are in larger enterprises (aka large clinics or hospitals), the unhappy docs will find anyway they can to get out of using it.

Why, you may ask, are these doctors so recalcitrant about technology?  The author said technology was not the problem.  According to him, docs use much more sophisticated technology everyday – MRI, cyro-probes, laser, etc.  Instead, it is because “It slows them down!”.   It isn’t the first time I hear this from a doc.  If you are in the business, I am sure you know this tune too.

Most EMRs, similar to ERP and other enterprise software, suffer from the same shortcomings.  Told poignantly by the author in this excerpt, the consumers suffer from a sign at a dry-cleaner’s shop reads: Low Prices, High Quality, Fast Service: Pick One. I am optimistic that we can get to a place where EMRs not only contribute to better care, lower costs and also user satisfaction (physicians’ and patients’).  It wouldn’t be trivial.  Some vendors won’t make it through.  If you have answers to the questions that the author raised in the article, I would love to see your comments.

Add to FacebookAdd to DiggAdd to Del.icio.usAdd to StumbleuponAdd to RedditAdd to BlinklistAdd to TwitterAdd to TechnoratiAdd to Yahoo BuzzAdd to Newsvine

The slow revolution in Healthcare IT

For years, healthcare IT has been dominated by big players and arcane system architecture that resembled the ERP market in the 1990s.  Playing into this is the vendor dominated CCHIT certification process, which required everything under the sky to pass, but did little to ensure the software support a better care delivery or even integration with the disparate systems used by different physicians.  However, there are reasons to be optimistic as 2009 comes to a close, as written in this blog.

Looking at healthcare IT, I see a lot of parallels to how the enterprise IT space has evolved over the last 15 years.  There were many promises and much money spent, but little to show in the form of tangible quality improvements or cost savings.  Sure the marketplace will eventually sort out winners and losers, but I postulate that government stimulus money will make the outcome of this evolution different.

Granted, what needs to happen in this space isn’t limited to just moving the power from traditional HIT vendors.  The interoperability in HIT in general is abysmal.  Even new vendors aren’t much better at allowing physicians and patients share data.  However, there are reasons to be optimistic that people building HIT software would consider a multi-tenant system or a data sharing platform a fundamental requirement instead of an afterthought.

What do you think?  I am especially interested in hearing from those who are involved in healthcare delivery.  After all, I am peddling wares just like everybody else.  You the user should have the final say.

Add to FacebookAdd to DiggAdd to Del.icio.usAdd to StumbleuponAdd to RedditAdd to BlinklistAdd to TwitterAdd to TechnoratiAdd to Yahoo BuzzAdd to Newsvine

Managing debt with Web 2.0

I took some time off from blogging to get started on a new project.  I am still far from fully ramped up in this new space.  However, that’s a story for another day.

This entry is about managing personal debt, especially the credit card kind.  I think we can agree that the American society is on average not very savvy financially.  Add that to the credit card companies and banks that chased profit without any ethical compunction, we have brought on ourselves a global financial crisis.

Regardless of what the government is doing around better disclosure by credit card companies, people really should know what kind of impact their credit card debt and other loans will have on their financial health long term.  For this, I think DebtGoal.com is a valuable resource and worth checking out.  As Newsweek pointed out in an article, you’ve got to Bail Yourself Out.

Some of the interesting facts that they might not have shared on their website:

  • These guys are not software engineers.  Rather they are former bankers who decided that there is a business opportunity in helping people save money.
  • They have not really spent their money on marketing.  Instead, this is really guerrilla marketing at its best.  You can see it by simply typing DebtGoal into your favorite search engine.

Given the recent acquisition of Mint.com by Intuit, I wonder if there are further web property consolidation in the personal finance space.  Obviously the credit card companies can generate some good will with congress by purchasing such technology.  At the same time, providing this kind of utility would reduce their profit since they have always count on you not knowing how much those cash advances and coffee are costing you.  A knowledgeable consumer is an unprofitable customer.

Add to FacebookAdd to DiggAdd to Del.icio.usAdd to StumbleuponAdd to RedditAdd to BlinklistAdd to TwitterAdd to TechnoratiAdd to FurlAdd to Newsvine

Zoho, an unassuming success story

I am late to the party on this one but hey, it is better late than never!  In talking to Sramana Mitra recently, the profilic writer and thought leader on entrepreneurship in Silicon Valley, I discovered Zoho.com and their whole slew of applications.

Apparently you can log in to Zoho.com using your Google Docs account.  They offer free access to a bigger collection of Office-type applications than Google, including mail, word processing, presentation and spreadsheet.  In addition, they also offer polling, web conferencing, CRM, etc for both personal and professional use (at a fee obviously).

What intrigued me the most is the story of how the founder, Sridhar Vembu, runs the company.  I have not met the guy personally but would love to some day, because he is showing the software business a different way to run the business and is very successful at doing it.

A few of the things I learned about him from my conversation with Sramana:

  1. Zoho (formerly known as AdventNet) is fully owned by Sridhar Vembu.  He grew the company organically.  VCs have approached him but he had no need to for their investment given the millions of dollars of cash flow the business generates.
  2. Vembu has over 1000 employees.  Many of whom he recruited directly from high school and trained into programmers in the university that the company runs.  He recruited from candidates pools that were traditionally underserved by local universities and got unwavering loyalty from those he offered the opportunities to.   He definitely has the guts to think differently and do it.
  3. With his success, Sridhar Vembu remains understated and does not call much attention to himself.  They have fewer than 10 employees in the San Francisco Bay Area.

As Microsoft starts to put part of their Office offering online and Google continues to push Google Apps, many people have actually been using Zoho’s product offering.  Here is another article from ReadWriteWeb titled Zoho: The Little Engine That Could (Take on Both Microsoft and Google).  These guys are certainly something to watch.   I suspect that Zoho might actually be making more money than Google Apps but I do not have any data to back up that speculation.

Add to FacebookAdd to DiggAdd to Del.icio.usAdd to StumbleuponAdd to RedditAdd to BlinklistAdd to TwitterAdd to TechnoratiAdd to FurlAdd to Newsvine

SaaS vendors who eat their own dog food

Should SaaS vendors be running a lean and mean IT group and utilize SaaS versions of ERP, CRM and other applications systems?  Mary Hayes Weier wrote on her InformationWeek blog that SuccessFactor has signed on to use NetSuite to run their ERP but Salesforce.com is still using on-premise Oracle applications to handle account receivables and payables.

I agree with Mary’s assessment that SaaS vendors should really be commended when they fully adopt SaaS software to run their own business operations.  After all, there are a few concerns with SaaS that this practice would help eliminate:

  1. SaaS still does not scale for a large corporation with huge amount of data.  It seemed that Salesforce.com may actually be struggling with this issue with their A/R and A/P system.   If SaaS companies can’t figure out how to solve this problem, then their own growth projections are in question as well
  2. SaaS is not secure enough.  Data hosted at the vendor can be leaked or  stolen.  I don’t believe mature SaaS vendors is less secure than many of the corporate IT infrastructure, but if a SaaS vendor can’t trust another trust vendor to run their business applications, we may not want to trust either of them to run our business applications.
  3. Complexity in integrating SaaS with other applications.  Vendors will  tell prospects that they have solved the integration issue.   Data exchange between SaaS and in-house applications can be taken care of.  However, if the SaaS vendors themselves can’t show that they can rely on such integration in-house, you may not want to believe their sales pitch either.
  4. Long term cost effectiveness of SaaS vs on-premise software.  There are economic models that show SaaS applications are cheaper to run in the short run, e.g. for 3 or 5 years but if you extend the payments out to 10 years or more, SaaS actually costs more than on-premise.  By using SaaS applications to run their own business, vendors of SaaS can show that using SaaS for the long haul is a sound business decision and they are going betting their own success on it.

I think these are in fact questions that you can ask your SaaS vendors.  The point is not to make your favorite salesperson squirm but it is to force the question that if I were to run my business with your application, can you concretely show that your company believes in the economic value of your solution?

Add to FacebookAdd to DiggAdd to Del.icio.usAdd to StumbleuponAdd to RedditAdd to BlinklistAdd to TwitterAdd to TechnoratiAdd to FurlAdd to Newsvine