This weekend, the Texas court system rolled out a new website design for the Texas Supreme Court. The old “supreme.courts.state.tx.us” domain has been retired, with the Court’s information now folded into the larger “txcourts.gov” domain that already handles the courts of appeals. The new page format is cleaner and more modern. It will be good for humans with taste.

The software scrapers that feed this blog are, unfortunately, still trained for the old website. I’ve shut down the scrapers for a short bit until I can re-train and re-test that software. That should freeze the data in place as things stand now.¹

A compressed work schedule will prevent me from opening the code editor this week. My best estimate is that the blog will return to normal operation sometime the week of October 27th or the week of November 3rd.

Update 10:00am: Among the items that relocated were the Court’s previous slip opinions. Links that formerly went to a specific opinion now go to the Court’s main page.

I’ve written before about how entangled this blog is with my docket-tracking software. One side benefit is that, when there is some major change to the Court website, I can implement a change in one place and have the fix ripple through the blog archive. (This is not my first rodeo.)

Today’s payoff is this: I’ve written one quick fix for the opinion links in recent blog posts and some data tables. Those should now take you to my archived copy of the slip opinions, rather than to the Court website itself.² This fix doesn’t make the scrapers any smarter. But it does make the blog a little more usable in the meantime.

If you subscribe to SCOTXblog by email, you’ll see a new format today, as well as a new delivery schedule. Going forward, the blog is using a new email provider. The update emails will be sent weekly on Monday mornings, collecting posts from the previous week. The format should be much more pleasant, the burden on your inbox lighter, and the service more reliable.

If you (like me) also like quick updates during the week, I encourage you to check out the blog’s RSS feed or Twitter feed. Those are continuing as before. And you’re of course always welcome to stop by the blog’s webpage.

Whenever I hit “Publish,” I’m very conscious that my post is about to be read by a unique group of Texas attorneys and others interested in the courts. Thanks again for reading and subscribing.

SCOTXblog will be undergoing some major construction this summer.

The timing is driven by the imminent launch of TAMES in the Texas Supreme Court. The Houston courts have already launched the system, and so I have a better sense of how these changes will affect the websites that I’ve built to follow the Court.

As you’d expect, I will be watching the TAMES rollout as closely as anyone. It will include many of the features I built for DocketDB — along with some others that no one outside of the courts themselves could have implemented (like copies of motions and letters).

They call the internet a “web” for a reason

Here’s a rough sketch of how my Texas court-watching projects fit together:

On this chart, the blog you’re reading is near the middle. My personal court-tracking database is at the bottom, and you can see how some of those connections flow through to public sites (like the blog or the public areas of DocketDB).

As crazy as this chart looks, it seems to work. Some of the scrapers have been running for more than five years with very little modification.

The future of DocketDB and SCOTXblog

While TAMES is a great deal for the public, it has a cost for me. My private court-tracking system is built on a foundation of “scrapers”, tuned to understand the old website. When the new site goes live, the Court’s docket information will be shifted to new pages and put in new formats. The existing scrapers will not know how to find or work with this information, so data will stop flowing into my private database, which in turn will freeze the public version of DocketDB. (( This has happened on a smaller scale at various times this spring, as the Court has made small improvements to its website. Even tiny changes on the Texas Supreme Court website — such as putting the description for a particular kind of docket-sheet entry into plainer language — confuses the scrapers. I’ve been able to make small fixes as these have come to my attention. The TAMES rollout will be an order of magnitude (or two) more complex. ))

For that reason, I anticipate that the public version of DocketDB (including registered accounts) will also wind down soon after the Court switches over to TAMES. That site needs a consistent flow of accurate data, and I won’t have that until I build a new set of scrapers for my own tracking system.

My first priority is fixing SCOTXblog. The easiest approach would be to cut the connections between the blog and my court-tracking software, leaving just a basic blog in place. But that’s not my style.

I’d much rather build a new platform for SCOTXblog that can better support the kind of case updates and articles that I’d like to write — and that, I hope, you’ll want to read.

I’ll update you as I make progress. In the meantime, you may see some longtime blog features start to disappear or stop working (like the feed of news articles or briefing requests). If you see something that looks more broken than usual, please let me know. And, of course, please stay tuned. I look forward to sharing the new site once it’s ready.

You may have seen the “stat pack” offered by SCOTUSblog with voting statistics for the Justices.

Its wedge-shaped voting chart is familiar — and, for a time, it became a staple of Texas CLE programs as well. Following that path, I prepared some of these charts for an Austin appellate bar program in March.

The experience made me more conscious of two weaknesses in these charts.

First, the author often must choose what universe of cases to show — all opinions, all signed opinions, only divided opinions. The easy choice is “all opinions,” but it gives very little signal compared to noise. In other words, the unanimous opinions push up the percentages and hide disagreements in the more challenging cases.

Second, even with a background in economics and math, I was bothered just seeing a percentage showing that two Justices agreed 60% of the time. That top-level statistic is fine for people who just want to make a broad-brushed assertion about the general makeup of the Court. But as an advocate I wanted to know which cases led to that disagreement. That is where a serious advocate can start to learn how two Justices might approach a problem differently.

Announcing DocketDB voting tables for 2010

I finally have a set of voting charts that make me happy. The basic layout should look familiar.

But this chart is dynamic — you can choose to focus on the scope of cases you want (all opinions? only signed opinions? only those with a dissent?):

You can also choose the types of disagreement you want to highlight (disagreements in opinion? or only in judgment?):

And the real kicker for me: You can click on a cell in the table and drill down to a page that shows — for each pair of Justices — the specific cases on which they agreed in the opinions, agreed in the judgment, or did not agree in the judgment.

These detail pages show keywords about each case (when they are present in DocketDB). You can also navigate directly from here to each relevant opinion that led to the split.

Where to find the voting charts

The dynamic 2010 Texas Supreme Court voting analysis charts are available to anyone who has registered for DocketDB. (If you registered before and have forgotten your login, you can get access through this link. If you haven’t registered, you can register here.)

Once you’ve logged in, look under your “Research Tools” menu in the top left of the menu bar. You should see the new “Vote Analysis” option.

I’ve always reserved some of the DocketDB features for the professional subscribers who help support the site financially. With this set of voting tables, I have reserved the data for 2006 to 2009 for those supporters.

For the future…

I haven’t yet decided on a next step, but I’ll mention two further enhancements that could be built on this same set of data:

1. Choosing cases by subject matter, rather than a whole term. Back in March, I was asked if it was possible to look just at the cases involving a particular issue. The answer now is that, yes, it is possible. A chart can be made for any set of cases, defined on any characteristic in the database. (The trick is defining those sets of cases, or creating a tool that lets users create their own sets of cases.)

2. Graphic representation of voting patterns. I’m thinking of the slides that I presented in March 2010, showing the relative voting position of the Justices in 5-4 cases. (Here’s the slide (PDF) I made for the U.S. Supreme Court, to give you an idea.) The data that powers today’s dynamic voting tables should be able to power those same graphics. (I used a different computer to build those charts, and the trick here would be folding this into the main DocketDB code in a way that works for the web.)

Those two enhancements could be combined to show, for example, a graphic representation in how the Justices vote on cases about statutory construction, or civil procedure, or medical-malpractice cases.

As always, I very much welcome your input about what new features would be helpful to your practice.

As regular readers know, the Texas Supreme Court is on its summer break. The fall calendar is set, and the next conference of the Justices is not scheduled until mid-August.

With my SCOTXblog time, I’ve been working on a new project that I hope to launch by the end of the summer. Yes, it’s also related to the Texas Supreme Court; I will announce it on the blog when it’s ready.

Along the way, I’ve been learning a few programming tricks. With one of them, I’ve updated SCOTXblog so that its layout rearranges as you make your browser window narrower or wider.

But these improvements do not, unfortunately, show up in Internet Explorer. Read more about the new design

I’ve set up a new Archives Page that displays all of the blog’s previous posts by date or by category.

Whenever you want to scan our older posts, the “Archives” link can be found on the right-hand side of the menu bar.

I’ve just moved the blog to a new hosting provider. In theory, everything should work just the same. If you experience any problems with the website or your RSS feeds over the next few days, please let me know.

Some of you may have tuned in on Friday to see my coverage of last week’s extensive order list.

I left town very early Friday for a family emergency. I’m still needed there, and with some case deadlines looming next week, I may not be posting for just a little while.

If there’s something you think the blog should cover on my return, please drop me a note.