Changes for Better (and Worse?): Part II

Earlier this week, I started blogging about how much things have changed for us in scientific computing and laboratory informatics, over the years, not always for the better. That first post talked about how much some tools have changed in this post:
http://outonalims.com/2011/01/24/changes-for-better-and-worse-part-i/
Now, here are some other aspects of these changes.

More Sophisticated Systems
Also going back to my post-undergraduate time around 20 years ago, I’ll give an example of the LIMS that I was working with in my job as LIMS Administrator/Lead Developer. My company was in its third generation of home-grown LIMS and wanted to buy a system. There were just a handful of options, some just too expensive for our budget, none with the major features we needed, so we purchased a LIMS brand that allowed us to purchase a copy of the code. We internally developed our own report writer, calculation tools, quality control testing features, and a number of other things for the system.

One option the system did not have, however, and that we strongly wanted the LIMS company to develop was an editing function. Yes, I do really mean that the users could enter data but not change it. To make a change, a user would phone up the LIMS System Admin (me) and tell them what they needed changed. At the time, few systems were built on commercial databases and the one we purchased was no exception. As such, there was a “back door” feature where I could get into the custom file system and make the change using hexadecimal (base 16) (yes, that was awful for everyone involved and time-consuming). Neither my users nor I felt this was “user friendly” and each time I went to the user group meetings, I would put this feature on the “wish list.” Each time, no indication was made that this feature would ever be offered and I was told to “tell your users to type more carefully.” Indeed!! Can you imagine that would be acceptable, today?!?! So, our systems have improved, feature-wise.

Unfortunately, as we create more systems and they are increasingly complicated, integration of these systems remains a challenge that some customers find overwhelming.

Things That Haven’t Changed As Much as We Think
Twenty years ago, we used to have programming tools that allowed us to “mark up” our text (such as RunOff which allowed us to print documents that had bold, underline, italics and other such now-primitive features). We still use markup tools, such as XML and HTML. We now have yet more sophisticated ways to use these types of tools, but they are still basically just ways of marking-up our files in order to use the data within them in some way.

When I want to calculate something casual and fast (like distances, fees, other day-to-day numbers), I still use division approximation in my head or do division on paper as it’s still faster than opening up MS Excel or anything else and typing it all in. Or, for something more complicated, I still use my old HP RPN (Reverse Polish Notation) calculator (uses machine order, not algebraic order). I only take the time to use something like MS Excel when I have columns and complex calculations and/or want to save it and/or make it look “nice” (relatively-speaking, I mean).

Instead of carrying heavy papers or manuals (customers always had a set of manuals, so I hardly ever carried them around), but we now carry heavy laptops around, which we have to take out of their case to be checked at the airport, usually. And if we drop them, they break, although unbound papers will scatter so that wasn’t necessarily always better. But we are still killing our backs, just in other ways than before.

Clearly Better
When we had fewer choices, we had much more work to do to force those systems to be what we needed them to be. While there is still a tendency to do way too much customization and configuration of systems on our still-huge projects, we do get much more of the important features up-front with the advent of more of the specific systems. To give some examples: we have report writers we can buy instead of writing programs to create customer reports; we can buy LIMS and ELNs that have drug metabolism-based features and pharmacokinetic calculations, rather than writing programs into the systems we buy; and, we can buy an ELN instead of going to conferences watching speakers talk about how we’ll be able to do so “any year, now.”

Bottom Line
We only make things “better” if we take on the actual improvements we’re given and to find ways to overcome the things that we’ve made worse. We should never accept that “change for the sake of change” improves things without applying careful thought to the situation.

Gloria Metrick
GeoMetrick Enterprises
http://www.GeoMetrick.com/