Taken straight from the SP2 documentation:
“Report model generation from Oracle data sources that run on version 9.2.0.3 or later is supported. You can generate Oracle-based models by using Report Manager, Management Studio, or Model Designer.”
This is huge and wonderful news. We were connecting SQL Server to Oracle, and then creating models off of the SQL Server tables to accomplish this. Needless to say, we didn’t do too much of this. And perhaps more needless to say, we will be using the feature heavily. I didn’t like that it was not included in the original release of SSRS 2005, but I guess it is one of those things you really appreciate more since you didn’t have it in the first place. Now to test how well it works…I’m assuming it’ll be fine like most everything we have experienced with SSRS.
In some cases, I just don’t like the field-by-field logging when creating an audit trail for my database tables. While it might be needed in some cases (and it might make sense), most of the time, I am find that a simple audit table that stores an entire record of fact data is more useful and easier to manage for reporting, etc.
So, instead of a generic audit table that captures all old and new values for many tables, I prefer creating an audit table for each of my important fact tables. Drop me a line if you are struggling with how to implement auditing (on Oracle, SQL Server, or MySQL) and we can chat.
I’ve got to say that tag-based filtering of data on the web might just be near the top of the “next big thing” list. I see people and websites leveraging it all over the place. CompUSA now uses tag-based filtering on their site along with other big names. I actually think it’s pretty cool and user-friendly……….or do I?
Cool? Yes. User-friendly? At first it might seem as such, but when I really want to do something tag-based filtering won’t allow, I get very annoyed. For example, let’s say I’m browsing laptops at CompUSA, and I want to pick any that have AMD processors as well as 15.4″ OR 17″ monitors. Tag-based filtering will not allow this. I pick 15.4″, and then I have to go back to pick and view the 17″ monitors separately.
What happened to simple, tabular reporting with multi-select? What was wrong with that? I could pick whichever options I wanted, it took up less screen real estate, and it required much fewer clicks. I’ve even played with tag-based filtering on a few web applications and web sites. It is definitely cool and simple, and if you know that filtering will be a straightforward and easy task for your users, it might even be a good choice. And, chances are, if you didn’t use any reasonable means of filtering data, tag-based might be a great way to go. But, be weary that it might spark your users’ interest like it did mine and make them wonder why the filtering couldn’t be just as powerful and easy as it was in the ”old” days.
I remember a time in enterprise technology when developers, albeit a small percentage, were “cowboys” – you know, people who “shoot from the hip”. I’m not talking about the irresponsible, incompetent people, but instead the people who knew what needed to be done, knew how to do it, and wouldn’t wait for other people to make it happen. Yes, a bit uncontrolled, but insanely, insanely productive.
I can’t help but feel an overarching culture shift in technology from ROI to security. This special “security beats ROI” mantra was always reserved for the most uncool, slow, unproductive organizations (and unfortunately, it seems government always got the brunt of that since they were never driven by profitability). Well ironically, the trend seems to be reversing. Many areas in government seem to be taking a customer-focused, run-our-organization-like-a-business approach, while some businesses are taking a do-it-securely-and-perfectly-even-if-profitability-is-at-risk approach. Now, I am not saying that either is right or wrong. There are certainly reasons to focus on profitability that need not be explained. And the argument to focus on security and risk mitigation is also valid that need only be explained in one word: Enron.
But, I guess the part that I miss is that rogue developer who would come up with something cool and amazing [in the enterprise] every few months just because he wanted to make an impact. You know where all these developers have gone? Google and small businesses/startups. Google makes everything a beta product, they charge the public for almost nothing, and most importantly, they make enough money in their advertising that they can splurge on cowboy developers. Thus, you can be a developer at Google and work for years without touching anything that would be considered a material risk to the business. It is no surprise that they are attracting so many of the smartest. Part of me hopes that the security-beats-ROI trend in the enterprise is balanced back just a little so we can see our cowboys flourish again.
It is no secret that [very good] IT resources are a rare breed in the enterprise IT world. But, what is not apparent to many managers, recruiters, and executives is “Why?”.
All-too-often, developers, admins, architects, and higher level decision makers get caught up in the enterprise culture, a culture where each person is given the smallest piece of a bigger picture, where each person focuses on not just one business area, not just one application, not even just one aspect of implementing an application, but instead, people are expected to spend an enormous amount of time on one small – actually, tiny – piece of an application.
I was asked to work with someone who was a Senior Java Developer from a large enterprise and was shocked at what I experienced. This person could not form a simple SQL select statement to pull data out of an Oracle database. It wasn’t enough that I was shocked, however. This person themselves was shocked that developers at a mid-size corporation were actually expected to know how to design, build, and maintain an entire Java-based application (even if just a small one), including how it interacted with the database. I learned that she had come from an environment where she owned a single Java class and maintained it for years (it was pretty much her entire existence). It was a piece of code based entirely on Java logic and had no database interaction at all; learning how to work with a database was not going to be a problem for her, but learning how to do it effectively most certainly was.
What is your experience with developers that come from the largest of enterprises? I’m sure it has much to do with the specific organization, business area, application, team, and even direct manager, but my guess is that you are also finding technical people becoming more and more specialized (to the point of pigeon-holed) as technology becomes a larger “business” at every organization.
