Saturday, February 16, 2013

State of NEXLAB Address

My fellow weather enthusiasts, met geeks, forecasters, nowcasters, storm chasers (hopefully the respectable lot), and various other sorts of people who just so happen to frequent

Let's get down to business.

A couple weeks ago I launched all of our new analysis data, complete with redesigned pages, layout and new products to boot. This includes the birth of our very own RAP Mesoanalysis data and a vastly improved page for accessing a complete and customized set of raw, un-analyzed plotting maps (upper air + sfc + composite) all in a single PDF. On top of this I gave our navigational menus a cosmetic touch-up, bringing in some more appropriate color choices to help balance out the overall look of the site. As it so happened I managed to lump in some improvements to the models page such that now we can see the progress of the newest model runs of the NAM, GFS and UKMET. It's a minor improvement at this point, but the item that made it possible will mean some future improvements to our model comparison products... but more on that later.

Getting back to our analysis data; Barring a few touch-ups and some overlooked glitches to fix, I feel that area of the site is done for now. However mobile friendly versions of these pages do not exist yet. That is a priority at the moment. Our satellite and radar products don't seem to need much attention just yet. We have been having some problems with overlays coming up completely blank, which essentially ruins our meso-analysis feature for satellite. But to my knowledge, that problem has been fixed... hopefully for good. Our models page seems as good as it has ever been, yet as I said before there are some improvements I'm working on. The noticeable change that is to come will be some links that will make it VERY easy to jump to our model comparison products from any given model image. The rest of the improvements to the models page are really all behind-the-scenes work. Cleaning up code, moving some pages into more appropriate locations and... a few alterations to links... it has to happen.

With all of that said, I and the rest of the staff/faculty have been reflecting on the progress of the site since we launched "version 2.0" back in April of 2011. We feel like at this point the site is operating much closer to what we had originally been striving for. As such, we have been joking that the what we have now is "NEXLAB 2.5". That's nothing official, of course. However curiously enough, we have been batting around the phrase "NEXLAB 3.0". That concept has become less of an inside joke and more of an actual plan that we will follow through with... at some point.

To give a sense of what we expect for "3.0"; It all seems to be centered around 2 or 3 core elements. We want to re-work the design of the site such that it will favor larger and higher resolution images. Basically, we want to break out of the 800x600 box we've put ourselves in. We also are eagerly awaiting news from Bill Bellon and Tom Whittaker about HANIS (HTML5 Animation Software), the expected successor to Flanis and JSAni for creating animated loops of images; among other features. Once we have an idea of the capabilities of HANIS, that might actually be the point where we start moving forward with work for "3.0". The only other goal to achieve with "3.0" would be a fully tailored NEXLAB Mobile site.

But before we make that jump; I wonder if any of our users have been scratching their heads wondering when in the world we are going to give any attention to our text products. We've been avoiding working on them for a pretty simple reason; They work, they are exactly what everyone is used to and other areas of the site have needed more attention. Nevertheless it's the only untouched corner of our data site and it's time will come. However it is February and this time of year starts to become a whirlwind (quite literally) for our program. Our 2013 storm chases start at the end of April and there are many things to be taken care of before then. Then April through late-June we will all be occupied with running the trips. With that in mind we have opted to start work on text in July. Again, breaking ground on Text Products in July 2013. No ETA on launch.

Even before and after Text is started/finished, we have a lot of other plans to move on before "3.0" (again) becomes a realistic project. A project for our Models page that has slowly but steadily been progressing is an effort to switch our images from being generated by WXP to GEMPAK. We also expect to improve the coverage of our model "sectors" which will include at least 1, perhaps a couple western CONUS sectors. An exciting new idea I've had recently is to setup some user-defined "Global Settings" for the entire data site. For example; You may have noticed that we are making wide use of image "rollovers". In order to make these rollovers function fluidly on the site, we pre-load all of the images on page-load rather than having each image download "on-rollover". This means that a good chunk of data is being downloaded all at once. This (as we are aware of ourselves) is not always a desirable thing with a mobile connection that is touch-n-go out in the plains. A global setting I'd like to implement is a way to disable pre-loaded rollovers site-wide. Other settings along these lines will likely be included as well. I'd also like to make it just a "smidge" easier to "bookmark" specific sections of our data site. It's been brought to my attention that that would be desired among some users.

Continuing a laundry list of plans and schemes; We've also been talking about adding a handful of Terminal Doppler Radar sites to our single site radar page. On a related note; we've been expecting a pair of brand new servers to replace our existing ones (hopefully before April, although, no ETA on that). With these powerful new servers and increased overhead, we've been hoping to push them as much as we can. This would include upgrading a few more NEXRAD sites to feature full looping for all products.

I believe that's just about all we have planned at the moment. Although I must admit, now that I have it all written down it does seem like an ambitious and daunting workload. Nevertheless, these are our plans and now you know as much as I do. As always, feel free to comment here, our feedback page, or tweet us @CODMeteorology or @CoDWXData.


  1. I'm hoping you'll see this comment - couldn't find a direct email for you...I just want to let you know that I've loved your model page for years, but with today's updates to include 925mb and the numerous additional parameters you've added, it's absolutely incredible. I'm a Cornell Meteorology a TV Weather Guy in Boston...and adore your site. THANK YOU for making such a positive difference for our science, and for putting together a page that is so user-friendly, accessible and packed with data. Tremendous job!!!! All My Best, Matt Noyes

    1. I don't think I can express how appreciative I am of your comments, Matt. That means a lot, thank you very much! We all work very hard at this site. Our whole goal is to create a positive and comprehensive user experience. We're always working on improvements, so feedback is always(!) appreciated. Thanks again, Matt.

  2. Since when is adding 'INSHA'ALLAH' on the weather page acceptable?