In a recent post, Donovan Colbert discussed his vision for the future of technology, which included hybrid computing. Donovan elaborates about this topic from his professional experience in the trenches.
I recently wrote about hybrid computing, and the following post came directly from one of my comments in that discussion thread.
I work for a healthcare solution provider. This is going to be complicated, and I've got to be a little careful – but... here goes:
We host Electronic Practice Management and Electronic Medical Record services (EPM/EMR) for small, medium, and large medical practices. We also do billing, posting, and other medical clerical, administrative, and related services.
The federal government has mandated that practice adopt EPM and EMR solutions for medical records. They offer incentives for early adopters and fine those who drag their feet. There are also all kinds of regulations, including HIPAA and Red-Flag laws that affect the healthcare industry.
A small medical practice can't afford to host its own SQL DB, electronic prescription service, and EMR and EPM solution.
It really is a disaster and a tempest all at once. The doctors are doctors — not businessmen and certainly not IT. But what happens in an industry like this is that you get some group of doctors who either fund developers or learn how to develop themselves, and then they create an application and decide to market it. You end up with a mix of all the problems of healthcare and all of the problems of IT.
People in IT love to feel under-appreciated and think and/or ask, "If IT shut down, packed up, and went home, how long could this business operate?" And often it’s true. IT is seen as a cost-center, difficult to deal with, egotistical, and not in tune with the rest of the business. Also, IT is seen as being "enabled" to have this kind of attitude, because it’s critical and you're going to have to pay up for IT one way or another.
However, doctors are the core business commodity of healthcare. If they pack up and go home, it’s the same thing. They're highly educated and have pretty big egos. They can also be seen as abrasive, anal retentive, and marching around with a God complex.
This is a really bad mix. It isn't just oil and water – it’s oil, water, and a spark.
Now you've got doctors who've managed to avoid a lot of the IT world for far longer than other businesses. The cost of education has gone up, they have to constantly train on the latest advancements in medical knowledge and technology (another thing IT people complain about), and everyone wants a piece of their pie.
After doctors deal with huge student loans, staff salaries, medical equipment costs, and insurance hassles, the federal government comes along and says, “You've got to have EMR. You've got to have EPM. You've got to have eRx."
And suddenly the IT guys show up, and they say, “You're going to have to have SQL, and that takes a DBA. You're going to have to have workstations, and that takes desktop support. You're going to have to have networking, and that takes a network specialist. You'll need servers, and you'll need to figure out complex licensing. You'll need anti-virus. You'll need backup and disaster recovery. You'll also need UPS, cooling, and electricity. You'll need e-mail. You'll need e-Discovery. And you'll need scalability….”
Some doctors will find a consultant or consultant group that says, "We can do this for you." But a lot of times, a small, general consulting group isn't going to know enough about HIPAA or eDiscovery and how important that is in an industry where malpractice lawyers on retention swarm like flies and malpractice insurance is as necessary as car insurance. Not to mention, putting all that equipment in the medical office and having some small consultant support it when it’s mission critical clinical data you're dealing with (not dollars and cents lost in downtime, but potentially lives lost in downtime) – well, it just isn't a great idea.
But most practices can't have their own IT staff on site 24x7. IT operations is not their core business — healthcare is. Here’s the breakdown:
- Small healthcare providers: The small consultant is probably your only choice, and the doctor becomes his own IT expert. This is a disaster waiting to happen.
- Mid-size healthcare providers: If you’re in the middle, you're probably going to host, because you're too big to have a small consulting group nickle-and-diming you to death or to try doing it all yourself – and you're too small to justify servers and IT departments.
- Big healthcare providers: For larger operations, you might have a dedicated IT department, and maybe they absorb all of the necessary duties.
With the healthcare solution provider I work for, we host Practice Management and Medical Record solutions. That’s our core business. But we also provide e-mail hosting. It isn't our core business. And it distracts from our core focus, takes resources, time, and is generally outside of the scope of our mission.
Dealing with mail issues, such as e-Discovery and retention, drive me nuts. But there are lots of other issues. People have become accustomed to having GB of file storage in e-mail and being able to send huge files. But I don't have banks of SMTP and POP/Imap servers hooked to endless supplies of network pipe and SAN storage. My users can't understand a 50MB mailbox limit and a 3MB max file attachment size. Nor can they understand that when I increase sensitivity of spam filters, I get more false positives, and when I lower sensitivity, the flood gates open on Viagra and gambling ads. E-mail hosting creates a larger footprint for vulnerability on my network.
Even though we are a provider of hosted services, using a hosted e-mail service is very attractive for me. I spend more time and money for less return on e-mail than on any other app, platform, or service in my data center. I don't have the expertise to dedicate to e-mail that’s really necessary and can't afford that staff member. At some point, you're painted into a corner.
I've resisted the desire to outsource my e-mail needs for awhile. As an IT engineer, I fully understand the risks, liabilities, and compromises associated with outsourcing. But ultimately, I can't really see any alternative.
As for the "applications as a utility" model, I see it that way already. My licensing costs are a huge part of my operational overhead, and platforms like Exchange are not inexpensive. Microsoft knows this and really cranks up the price on their platforms that offer robust scalability – their "Enterprise" platforms. Standard platforms are priced for easy admission, but when you step up to Enterprise, you're going to get hammered.
Oddly enough, I’m moving all of my services and apps to a web-based model. I don't want fat client connections. I want to get rid of that model, because I end up having to deliver all kinds of kludges to enable my internal employees to work securely and effectively with external practices. VPNs, point to point T1s, shared documents and folders. There are lots of solutions where the industry is lagging.
Document collaboration and organization (think Sharepoint, intranet, and extranet) is really behind the curve, and critical. Business models have changed. In the past, a company's internal documents were generally only used internally, and if you needed to share, you would send them a floppy disk or fax it. Now companies are used to partnership-based collaboration. Companies are embracing remote workforces and work-from-home flexibility.
However, all of the solutions that allow your remote work experience to faithfully mimic your office work experience are complex and prone to issues that become an IT support nightmare. The more that I can make available through your browser, the less complicated my support becomes. It boils it down to this: “Can you get to Yahoo? No? Can you get to Google? No? Can you get to CNN? No? Ok... call your ISP." or if yes, "Ok... it looks like there is a problem on our end, let us research this and get back to you."
Right now, it’s – "Do you have a hardware VPN? Do you have a software VPN? Do you have a PtPT1? Do you know how to recycle your hardware VPN device? Do you know how to confirm that your software VPN has an active link? Do you have an internet connection? Is your home wireless working?" – and getting dragged into even deeper levels of support that are not really in our realm but that have to be fixed so that our services and apps are available.
So, from that perspective, I'm actually pushing to move our model to a "cloud delivery model" – to the point of almost removing the "hybrid" nature completely. By the time I'm done, I want all user documents published on some sort of searchable web interface, something like Sharepoint Server. I don't want file shares and word documents as flat files. That paradigm works best for single users on multi-user networks, not shared documents.
"I've got a shared folder, I'll copy a document here, you can pick it up, work on it, call me back, let me know you've put the modified version back, and I'll pick it up ". That works.
What doesn't work is, "Our department has 11 people who use this document, we collaborate with an external partner (with 11 people) who also uses this document… let's put it in a share, give them a VPN so they can see the share... now let's try to make sure we don't get multiple revisions of the document without any change tracking and oh well – someone accidentally deleted the file, anyhow… can you recover the last version from tape?"
The latter approach is dying, and a web-facing solution is the obvious answer. I'm not sure if Google Docs is going to be it. For me, sending that out to be hosted doesn't make sense, but I'll be selling it to other businesses where it does make sense for them to have it hosted.
The idea that hosting is going to so fundamentally change computing that everything will be hosted and you’ll no longer need Windows... that’s the Larry Ellison vision of the cloud, and I don't see that happening. I think that is pie-in-the-sky cloud optimism.