Its about the customer, not the technology

April 18 2008

Today, the hard work of multiple PSC employees on a discovery project was presented to a customer. We were reviewing an application framework and making recommendations. One of the pieces is the Output Generation Layer. The requirement is to have the output be in PowerPoint files. From there, they can generate a PDF or HTML, but we need something that in an electronic presentation format that scales.

When I came to PSC, I was a very one-sided consultant. I was passionate about Lotus Notes and Lotus SmartSuite. I tolerated Microsoft Office, because people used it. But anything else, I loathed. I was so focused on the holy war of one technology and brand that I hated everything else. It didn't matter if it was Microsoft, Novell, Oracle, or anyone else .. it was not Lotus. And for many years, only the Lotus brand at IBM was something I cared about.

PSC describes itself as a business technology services consulting firm. Our job is to listen (get it, it's all in the way we listen) and help solve problems. We believe in this listening foundation so much it is part of our email domain! Our job is not to force one technology on a customer. What we do is evaluate the problem, make recommendations, and then apply those recommendations to a technology the customer is comfortable with. We do get to make technology recommendations, and I am always thinking about how Notes and the other Lotus and IBM products can be used. PSC is bigger than just being an IBM partner. We have a large Microsoft team (and yes, we do SharePoint work ... lots of it) and we work with portals other than Websphere. We have a great open source team that is doing work with Django and other technologies. We have a health care practice in Kansas City. We even have SAP, Oracle, and Google practices within the company. This helps us stand out from many of our competitors.

I was asked recently "Can you be an evangelist and still be open minded to technologies you do not like?" I think that is a great question. I believe the answer is "Yes!"

So back to the customer proposal. Based on the needs of the customer, our recommendation is to build an Output Generation Layer that creates the PowerPoint files on the server without any interaction from the user. The best way to do this is using OOXML. I am going to let that sink in. Yes, we are making the recommendation to use OOXML. I spent time reviewing the functionality that allows me to create a .NET application and use OOXML to generate the PPTx files. It supports all the features we need TODAY in PowerPoint 2007. I do not have to wait for anything in the new version. Nothing from the standard's board as they begin the process of making OOXML a standard.

Now, I know some of you are out there scratching your head. Why would the guy who loves Lotus Symphony and OpenOffice.org and presenting a fiery presentation at ILUG 2008 make this recommendation? Because it solves the business problem.

Working with customers is not about fighting a holy war for one technology or another. It is about listening to them and solving their problems. It is about making accurate assessments and honest estimates. If your favorite technology solves the problem and the customer is comfortable with that, perfect. But holding on to a holy war does not solve those problems.

Maybe this is a sign that I am leaving the hard core development world (and for that matter, was I really ever there) and moving more into the business and sales side. Not sure, but I really am excited about this. Heck, maybe this project can become one of the first OOXML case studies for PowerPoint!

9 Responses to “Its about the customer, not the technology”

  1. 1) Khyle says:

    I started out as a Domino administrator. I generally liked the platform, although it required a fairly heavy commitment in order to be truly useful.

    The more gray hair I get, the less I like any given technology for it's own sake. At the end of the day, the strengths and weaknesses of any given technology are really just business decisions made by the owners of that technology.

    That's one reason MS is having trouble innovating, they feel the need to maximize revenue on everything they do. If they can't find direct and indirect revenue sources, they won't do the project.

  2. 2) Axel says:

    Couldn't agree more.

    And of course Microsoft does research where the roi is expected far in the future and as a result difficult to predict. Who does not believe, might listen to this episode of SE-Radio ({ Link }

    Also interoperability tends to become cheaper with all those xml, component frameworks, service buses, etc. The landscape is changing and everything like consulting, marketing, programming has to adapt to an ever changing world, anyway.

    Our customers are interested in solutions, not in coarse grained, ill-mannered tech debates loaden with prejudices. Those might help vendor sales and vendor marketing, but undermine our credibility as analysts, consultants, architects, designers, programmers, testers, journalists or whatever.

  3. 3) Andrew Lauter says:

    Wonderfully put. Its about listening and solving business problems. Way to go John.

  4. 4) David Bailey says:

    "...I am leaving the hard core development world..."

    Maybe you're selling your soul! :-)

    "...was I really ever there"

    Nobody was ever more there! :-|

  5. 5) TripleII says:

    I will ask you, which version of the "standard" do you implement. The Pre-BRM ECMA standard that most closely matches Office2007, the post BRM ratified standard that will function with no programs, or the to be continuously revised version under ISO control? This is a serious question. If you code to non compliant pre-BRM, you have nothing that can be certified but comes closest to working (good luck with that though, more questions than answers).

    The ONLY use OOXML has will be data extraction and many post processing to get your data OUT of the XML like structures at a later time. MS won't be compliant, but will mix and match SOME of what ISO adds/revises to keep their marketting stamp approved, but not enough for anyone else to actually implement it.

    See, therin lies the problem. ODF has been implemented successfully by 10+ vendors. What company is going to invest the 10x effort to implement OOXML when there is no way to validate it might work, temporarily, with some future version of office. We already see this, read only plug ins for Apple, IBM and now OpenOffice. Get the data out, but writing is just not going to work.

    TripleII

  6. 6) Nathan T. Freeman says:

    I want to talk to this customer in three years, and see if they still agree with your recommendation. I'll make you a personal wager of $100 that by Q2 of 2011, if they are still using this approach, they'll find it inadequate.

    Does that mean they shouldn't do it? Of course not. But the OOXML specification is a joke, and the resulting output they're going to get from it is laughable. It might do the job at the moment, but they'll run into an interoperability, maintenance, or scaling problem by 2011.

    This is EXACTLY like building a DHTML solution in 2000 that only worked with Internet Explorer, because the app-level tools on that platform really were ahead. I did it back then, too. By 2003, every site that relied on IE-only behavior was considered badly designed. And today, we find such web efforts downright laughable.

    Sometimes, John, you have to listen to the things that the customer *doesn't* say. Sometimes being a consultant means that you have to correct the customer's mistakes. After all, they hired you precisely because you understand this stuff better than they do. Or at least, you're supposed to.

    They're paying you to be smarter than them.

  7. 7) John Head says:

    I was waiting for you to reply Nathan ...

    First of all, I did listen to the customer. They customer didn't care how we did it, but they need a single content source that is printable and presentable by all of their customers. Without any resizing of the content by the end user. Without user intervention. And the output count of over 1300 files in a 48 hour period. There is no other way to do this then a server-side format, and that format is PowerPoint.

    Would you care if I said I was going to automate PPT on the server side? Would that offend everyone a little bit less? Automation on the server side would never scale.

    Second, we reviewed every technical solution available, including ODF. The ODF spec does not support all of the functionality we need in the charts. Put aside the religious and political battle that ODF vs OOXML has been the past year plus, this is about what can be done today. ODF fails.

    I am not worried about what will happen in 3 years. First, they could keep using PowerPoint 2007 if they wanted to. But most people who are ready to put a system like this in place know the risks of future software upgrades. A well architected and built system won't need a major overhaul to adjust to a new version of PowerPoint or PPT.

    The consultant pays me to solve their problems Nathan. I am doing that in spades. They are ecstatic. They are not attached to battles fought or ideology. They are tied to their business model and making a living.

  8. 8) TripleII says:

    Your final entry has nothing to do with OOXML and it's bennefits or defects, what you have centered around is a SPECIFIC application that suits your need, meaning PowerPoint. There is nothing wrong with the solution, but has nothing really to do with OOXML. Even though Office 2007 uses an output format LIKE OOXML

    { Link }

    Your blog entry about OOXML being about customers is not really valid. What you are actually talking about is simply that Office 2007 MEETS the needs of your current customers. The fact that Office uses OOXML LIKE storage or binary formats or stores the documents in compiled VB assembler is irrelevant, you needed a server side PPT solution, and Office worked for you.

    Again, that is a fine solution, however, it will require Office 2007 until the end of time to work flawlessly. Unless you believe (it has never happened before) that future versions of Office will have perfect backward compatibility. MS doesn't do perfect backward compatibility (it doesn't drive upgrades if Office 2010 renders Office 2007 files perfectly)

    TripleII

  9. 9) benha university says:

    I'm grateful for your visit and I hope to visit my website to share information

    { Link }

Leave a Reply