Image via Wikipedia

The Article…

Alan Cohen, VP of Marketing for Nicira, a disruptive startup focusing on altering the IT networking landscape, recently wrote a guest post for TechCrunch suggesting that an “Arab Spring” of sorts is occurring within IT. Reader comments have indicated a significant distaste for the analogous use of Arab Spring in comparison to IT – the one involving the loss of human lives, the other changes in business structure and Mr. Cohen has apologized for the offense some feel at the branding. I don’t think it was the best title for the post – but there has already been rampant discussion of that fact – and I would like to focus more on the content of Mr. Cohen’s post.

Cohen suggests we are seeing a seismic shift in the IT sector – similar to that which occurred in the 1970′s with the advent of the personal computer. He notes early innovators like the iPhone (antecedent to smartphones), VMWare (virtualization), and Salesforce (SaaS). While I agree with his historical evaluation of the situation and also that a seismic shift is occurring within the IT sector – I do not share his optimism about the nature of this change.

Cohen states, “If IT providers do not supply what the end users want, the latter, like
the brave individuals who took the streets of Cairo, Tunis, and Tripoli,
will take matters into their own hands.” He deems this “shadow IT:” “Bring your own device is shadow IT. Most SaaS applications start by
bypassing IT and going directly to functional groups (managing sales
through Salesforce or sharing through Box.net).”

Now, working outside the box with one’s own devices/services concerns me enough (e.g. smartphones, file sharing), but Cohen then suggests that it is time for revolutionary employees to step things up a notch – implementing their own infrastructure: “If IT does not provide the end user with the infrastructure they need,
the latter can rent it, by the hour or month from companies like
Rackspace or Amazon. All you need is a credit card and no approval from
IT.”

Cohen concludes by challenging those within the IT industry, “But if you are in IT, you have to ask yourself: What side of history will you wind up on?”

Problems with the Article…

I do work in IT. I’ve been a geek practically since I was born and have worked full-time in the industry for the past six years. But this isn’t about my IT credentials…and it shouldn’t be about defending IT as IT. Truth is we have our weaknesses. Many of us who work in IT are a bit antisocial, introverted, ADD, OCD, or (append acronyms here). We aren’t always the most flexible folks to work with and there is certainly room for growth in our lives and skills…That said, I do think there are some serious problems with the sort of revolution Cohen is endorsing here…and I think his IT guys probably grimaced reading this article as I much as I am now.

Lets talk for a few minutes about the dangers inherent in this sort of “revolution:”

Reduced Productivity – While using a device familiar to the end user may be a pleasure and expedite that individual’s productivity, it may result in a decrease to organizational productivity. For example, if one finds an Apple iPhone easier to utilize than the company provided Android or Blackberry devices (or whatever might be the standard) it is very likely that generally you will not require IT support in utilizing your phone…but then there come those times when you do…

• When the device craps out and refuses to boot.
• When you can’t get the company email/calendaring software to sync with your phone.
• When someone steals your phone and you really, really need to have that sensitive business data remote wiped.
• When you sell your phone on eBay and someone pulls sensitive data off the storage because deleting data isn’t enough.

At these times stress goes up and productivity goes down – for you and for the IT department. In time that could have been spent deploying five new computer systems only one thing is now accomplished – getting a non-standard piece of hardware working again (or wiped or whatever).

Reduced Security – There is significant danger in the use of non-standard equipment and services for achieving business goals. It is hard enough for IT to maintain complex password requirements on the variety of internal systems any organization utilizes, let alone managing password requirements on non-standard devices. This is especially true of file sharing services. A handy example would be the recent takedown of MegaUpload. While largely used for illegal purposes, numerous individuals used that site for personal or business purposes – to manage their files. It seemed “safe” – and now it isn’t. Doesn’t matter how hard your IT department tries – if you put your files on a system that is raided by the feds, say goodbye to that data – even if it is important financial info. for the management of your institution.

Reduced Effectiveness – Cohen greatly oversimplifies the simplicity of the services currently available to perform complex functions. Amazon, Rackspace, nor even Salesforce are “insert your credit card get your DVD rental” simple services. For the last few days I’ve been spending a decent number of hours pouring over books and documentation on Salesforce – and it isn’t a piece of cake. Do you want to set up Users? Roles? Profiles? Groups of Settings? Are you sure that the way you setup the permission inheritance will prevent Sally Jane from seeing everyone else’s social security numbers? Do you want to use Salesforce? Force? Heroku?

Valid Pain Points…

I understand there are real pain points for end users in organizations. Not being able to share files in a simple and efficient manner is frustrating. Using outdated and clumsy software to manage customer relationships is frustrating. Learning new technologies and devices outside your comfort zone is time consuming. These are real issues and IT needs to pay better attention to them…but…and this is a big BUT…looping around IT will not decrease but instead increase these pain points.

Sure, you might be able to happily use an iPhone even though IT says no and never have a single problem…but then again…and as the complexity of the device/service increases the issues exponentially increase.

Finding a Road Forward…

Now the real question is, what is a workable way forward? How can end users and IT cooperate to achieve optimal effectiveness, productivity, and security? Well, it isn’t an easy or short road…but then again, real revolutions never are. I’d suggest that the consumerization of IT is a short-term bandaid fix for the real needed change. Too often we take the easy way out rather than working through the difficult decisions that really need to be made.

• Communication - IT needs to get better at communicating what they are doing in their little forests of IT solitude and why we aren’t really just sitting around playing video games and watching youtube. On the other hand, we need end users to get better to actually listening to IT. So frequently folks ask me, “How did you do that?” But they usually don’t want me to tell them…b/c as soon as I open my mouth their eyes glaze over and their fingers drum impatiently on the desk. Now, I know I’m using some technical terms – but I’m willing to explain them if you are willing to ask questions and listen. Maybe with time you’ll learn more about the tech terms I use and I’ll get better at describing technology in clearer terms.
• Prioritization – One of the biggest challenges facing everyone everywhere is the constant call to do more simultaneously and faster. Oftentimes the issue for IT is not lack of desire to improve a given service to the end user, but lack of time. This means the organization as a whole and its individual departments need to determine what projects are most important – and what projects (in spite of being so important) are less important.
• Consideration – IT needs to consider not only what is the best technology but also what will have the greatest net positive effect for end users. Meanwhile, I’d recommend instead of bringing in new devices and services to circumvent IT folks ask IT, “what can I do to have the greatest net positive effect for you?” By focusing on those areas you reduce time IT has to spend in them and free IT to focus up on working on those devices and systems you really want.
• Forgiveness – We can all be nasty at times…and if you haven’t personally been nasty to someone, I guarantee someone else in your department has. A lot of us walk around with a heavy bundle of war wounds. You walk into an IT person’s office (or vice versa, into an end user’s office) and a single word, your position, or what happened earlier today may cause that person to dive underneath their desk as if someone just shouted that mortars were incoming. Getting along means letting bygones be bygones and when new items come up, dealing with grace and humility – for both sides.

What do you think? What is your experience working with IT folks? Or working with end users? Is consumerization of IT really the way forward? If not, what is the alternative?

Tonight I’m going to share one of those humorous stories – or at least I think it is humorous…though it was not so when it occurred to me earlier this evening.

I was working a late shift at Philadelphia Biblical University and performing some standard security patches and software updates. Earlier that day I had removed a chunk of unused space from a database server virtual machine. Everything had continued operating normally and I’d thought nothing of it…

Imagine my surprise when, upon rebooting to complete updates, I received a black screen with those tiny glowing white words, “Operating system could not be found.” These are not the words you desire to see – especially at 8:30 p.m. when you are thinking about going home and relaxing shortly. Visions of long-nights of yesteryear fill your mind.

It is at this moment that the first comical occurrence of the evening occurred. Microsoft decided that on this very night at this very moment it needed to give me an inescapable four minute count-down to reboot my system in order to complete windows updates! COME ON! I happened to be burning a new ISO of Windows Server 2008 EE R2 at that exact time and I watched in desperation as the four minutes chewed away more quickly than Windows was able to burn the new ISO to disc. Yes, at 99% Microsoft overrode my every whim and desire and rebooted the system – if I did need that disc I’d have to burn it all over again!

In any case, after the system came back, I did some googling – but it isn’t all that easy to google some topics. See, the problem is in making Google (or any search engine) understand what you want. For example, I really needed to information about the “Operating system could not be found” message where there was also information about VMWare‘s vSphere and information about deleting unused space. I found a few relevant articles – but none that exactly addressed my issue or provided a resolution.

I booted into Windows Server 2008 EE R2 recovery mode off the ISO and attempted to run a repair – but it didn’t see the drive. Great. I went into the command line just for kicks and giggles and typed in C:. Hmmm…That’s interesting. The C drive is showing all the databases which are not on the C: drive…

It was at this juncture I recognized the next humorous tidbit – VMWare seemed to have arbitrarily reorganized the drive letters. If the drives were supposed to be Drive 1 – C, Drive 2 – D, Drive 3 – E they now here more like Drive 1 – E, Drive 2 – C, Drive 3 – D. No, this doesn’t make sense to me either.<ref>That said, I think there probably is a logical reason, just one that I don’t understand. This didn’t seem like a bug in VMWare’s software, rather like a logical but obscure process that one doesn’t run into except in awkward situations such as this.</ref>

More anxious scrabbling. More googling. I booted into the BIOS and looked at the boot order. Ahh, yes, it was booting off of Drive 3, not Drive 1. I began slowly rotating the drives – each one in turn to be the primary. After a few tries Windows booted without issue.

My evening was redeemed! The moral of this story is that when you do something really stupid with computers – it oftentimes works. When you do something that seems mundane and simple – it oftentimes backfires. Woohoo!

Before leaving Collages.net I was working a lot with different virtualization technologies – and I love them – and believe they are the way of the future. So I researched various VPS providers over at WebHostingTalk. I finally decided on VPSLand, despite some bad reviews, because of their AMAZING prices. On 7/26/08 I purchased a Windows-EZ Value or Busines plan from them (I don’t recall which). In any case, at the minimum I received 1280 MB RAM 400 GB Bandwidth, and so on – running Windows Server 2003 64-bit. AWESOME!

Unfortunately, I had not had previous experience with Parallels/SWSoft’s Virtuozzo and have since decided I absolutely hate it. I had a number of issues over a period of a few weeks, but knew a few where Virtuozzo’s fault and the rest I decided to give VPSLand the benefit of the doubt on. For those who aren’t familiar with Virtuozzo, it works differently from most virtualization technologies. Rather than completely isolating the virtual OS instance it shares the instance across all slices. The practical result of this is that you can’t make modifications to the Windows core (e.g. apply Windows Updates). Call this the network engineer in me – but when I get a VPS I want full control to upgrade/patch/install/replace however I see fit.

Next step – research alternative VPS solutions. It seems a majority of VPS providers utilize either Virtuozzo or its open source companion, so it took me a while to find ones supporting Citrix XenServer, Microsoft Hyper-V, or VMWare’s ESX. In the end I decided to stay with VPSLand and use their Xen VPS.

Double the price. I purchased one of VPSLand’s WindowsXL-1024 Xen VPS plans. It included 1024 MB of RAM and 1000 GB bandwidth. Not bad, certainly a step up – though I would have taken less bandwidth and less HD space in exchange for a lower price. No matter, I’m now paying $79.99/mo. Well, for over double the price I expect some good service and a good system. Now lets follow the timeline:

• Sept. 3rd, 2008 – Order places for Xen VPS around 3 P.M. EST.
• Sept. 4th, 2008 – Account information received around 7:50 P.M. EST.
• Sept. 4th, 2008 – At 10:22 P.M. EST I submit a ticket reporting that while I can connect to the VPS I cannot ping out to any remote server (e.g. Google, Yahoo).
• Sept. 5th, 2008 – At 4:27 A.M. EST I receive a ticket reporting that this issue has been resolved. The support tech. informs me not to apply Windows Updates, that they take care of this on their VPS.
• Sept. 5th, 2008 – At 11:09 A.M. EST I send a response asking why I can’t make Windows Updates since this is a Xen VPS not a Virtuozzo VPS. Also asking for an explanation on what caused the technical issue with my VPS.
• Sept. 6th, 2008 – At 9:52 A.M. EST I receive a response from a different technician that yes, the first technician was incorrect and I can install patches and then a cryptic answer to my question about the cause/resolution of the issue in the first place (something about restarting a network service on my VPS – strange since I rebooted the entire server multiple times).
• Sept. 6th, 2008 – At 12:35 P.M. EST I submit a new ticket. I’ve just noticed that my VPS is down – entirely. No ability to visit the sites, no ability to RDC into the server. No ability to access the web based control panel. This is a complete service outage.
• Sept. 6th, 2008 – At 2:41 P.M. EST I receive a response to my ticket – over two hours later. They are “escalating” my ticket to the “reboot queue.” Two hours and they are just now recycling the server?
• Sept. 6th, 2008 – At 3:47 P.M. EST I write again on the same ticket, “I am still down. The server came up briefly according to a pinging service but went back down again shortly thereafter. Please give me a status update.”
• Sept. 6th, 2008 – At 11:10 P.M. EST I wrote again on the same ticket, “I would love to get an occasional update – say once every two hours or so to let me know what is happening. Right now it kind of feels like you guys just gave up on fixing my VPS. This has been a horrible starting experience with VPSLand. Please inform me of the status – I am still down – nearly twelve hours after initially reporting this outage. How can I even place a minor website on a server that is going to be this unstable? And where support does not respond?”
• Sept. 7th, 2008 – At 3:37 P.M. EST as I write this post my VPS is still down, I have received no further communication from VPSLand and it has been well over 24 hours since my VPS went down.
• UPDATE: Sept. 7th, 2008 – I submit a new ticket in a desperate attempt for attention. Title is, “SERVER DOWN FOR 24+ HOURS!“
• UPDATE: Sept. 7th, 2008 – At 9:23 P.M. EST receive a response to my original ticket that this issue is being escalated to their senior admins (after well-over 24 hours straight downtime) and that they will give me status updates.
• UPDATE: Sept. 8th, 2008 – At 12:01 A.M. EST receive a note informing me VPS has been restored, also that they are giving me a credit for one month free service and apologize for the delay. Issue report is a failed drive (this is why one uses RAID – at least 1, perhaps 5 or 6).
  

At this juncture I feel the neglect is simply insane. “VPSLAND.com’s Virtual Private Servers are perfect for Businesses or Individuals looking for an affordable dedicated server alternative with full Root/Administrator privileges.” I’m running hobby sites folks – and this isn’t working for me?

So here are my thoughts at this juncture:

1. Talk to Us. People are very understanding when they know that there is a problem and that a company is working to resolve this problem. My problem is I don’t know if anyone is even working on this problem. Sure, the ticket is open – but is anyone home?
2. Better SLA. I should have looked at their SLA more closely. While they offer a 99.9% uptime guarantee their SLA more carefully defines this. If you have less than 99.9% in a single month they compensate at 100% if it is 89.9% or below – that’s an astonishing 72+/- hours in a single month (yes, that was the sound of my jaw hitting the floor). The SLA should look like: Below 98% 100% credit plus a 100% refund (e.g. you pay me).
3. How Ya Gonna Change? Why does it take hours to get a reply to a ticket? Why is there no phone number to call in emergencies? Do you need to hire more techs.? Do so. Jump the prices to do so? Okay. But this is unbelievable. Automatic migrations? Manual migrations? A must!

Maybe something at this point will change…and if it does I will inform you. But at this juncture – my personal opinion is – don’t ever EVER ever utilize VPSLand for anything (yah, grandma, not even your hobby site on knitting!).

Anyone out there have a good Windows VPS service (not using Virtuozzo) that wants a new client? I’m good PR if you treat me well (and bad if you don’t).

P.S. I’m softer on companies that haven’t been in business for a long time (e.g. startups). I know the difficulties experienced, but VPSLand or its antecedent have been running since 1994.