Get early access to
our products
The Sea of Tabs
Jun 6, 2023
When I talk to litigators about what we're building at midpage, I always talk about the sea of tabs. Tell me if this sounds familiar.
As you read Case A, Case B grabs your attention. You open it up in a new tab. Case C and Case D also seem helpful, so those get new tabs, too.
Halfway through Case A, you hit a jackpot-of-a-string-cite buried deep within a footnote. More tabs for Case E through Case J.
When you realize Case A has an unfavorable ruling, you switch over to Case B.
Case B, it turns out, is also a loser. But it quotes Case E through Case Q, and those sound promising. More tabs.
You might have opened up Case E already, but how the hell are you supposed to know? The tabs have gotten so small that you can't read the case names. Better safe than sorry. Now you have Case E open in two separate tabs. Before long, you have approximately 1,000,000 cases open in 1,000,001 tabs.
The plot thickens when you think you can contain it. You break your single window of tabs into multiple windows of tabs: Good cases here, bad ones there. You convinced yourself that it was, in fact, a good idea to add this extra layer of complexity to the hellscape of caselaw you created. Why limit yourself to one sea of tabs when I can have several?
If all of this sounds crazy, that’s because it is crazy. It is an objectively bad way to do legal research. But when I describe it to other litigators, they nod along in solidarity.
This is the kind of thing we have been thinking about as we develop our legal research and drafting platform. How can we simplify?
Yes, we leverage very good AI—including the generative variety everyone is so excited about. But AI is a means to an end, not the end itself. The end is a simplified legal research and writing process. That's it.
Next week, I think I'll write about another thing I despise about legal research: Putting it all together into an email someone will actually read. We're going to simplify this, too. What else should we fix?
When I talk to litigators about what we're building at midpage, I always talk about the sea of tabs. Tell me if this sounds familiar.
As you read Case A, Case B grabs your attention. You open it up in a new tab. Case C and Case D also seem helpful, so those get new tabs, too.
Halfway through Case A, you hit a jackpot-of-a-string-cite buried deep within a footnote. More tabs for Case E through Case J.
When you realize Case A has an unfavorable ruling, you switch over to Case B.
Case B, it turns out, is also a loser. But it quotes Case E through Case Q, and those sound promising. More tabs.
You might have opened up Case E already, but how the hell are you supposed to know? The tabs have gotten so small that you can't read the case names. Better safe than sorry. Now you have Case E open in two separate tabs. Before long, you have approximately 1,000,000 cases open in 1,000,001 tabs.
The plot thickens when you think you can contain it. You break your single window of tabs into multiple windows of tabs: Good cases here, bad ones there. You convinced yourself that it was, in fact, a good idea to add this extra layer of complexity to the hellscape of caselaw you created. Why limit yourself to one sea of tabs when I can have several?
If all of this sounds crazy, that’s because it is crazy. It is an objectively bad way to do legal research. But when I describe it to other litigators, they nod along in solidarity.
This is the kind of thing we have been thinking about as we develop our legal research and drafting platform. How can we simplify?
Yes, we leverage very good AI—including the generative variety everyone is so excited about. But AI is a means to an end, not the end itself. The end is a simplified legal research and writing process. That's it.
Next week, I think I'll write about another thing I despise about legal research: Putting it all together into an email someone will actually read. We're going to simplify this, too. What else should we fix?
When I talk to litigators about what we're building at midpage, I always talk about the sea of tabs. Tell me if this sounds familiar.
As you read Case A, Case B grabs your attention. You open it up in a new tab. Case C and Case D also seem helpful, so those get new tabs, too.
Halfway through Case A, you hit a jackpot-of-a-string-cite buried deep within a footnote. More tabs for Case E through Case J.
When you realize Case A has an unfavorable ruling, you switch over to Case B.
Case B, it turns out, is also a loser. But it quotes Case E through Case Q, and those sound promising. More tabs.
You might have opened up Case E already, but how the hell are you supposed to know? The tabs have gotten so small that you can't read the case names. Better safe than sorry. Now you have Case E open in two separate tabs. Before long, you have approximately 1,000,000 cases open in 1,000,001 tabs.
The plot thickens when you think you can contain it. You break your single window of tabs into multiple windows of tabs: Good cases here, bad ones there. You convinced yourself that it was, in fact, a good idea to add this extra layer of complexity to the hellscape of caselaw you created. Why limit yourself to one sea of tabs when I can have several?
If all of this sounds crazy, that’s because it is crazy. It is an objectively bad way to do legal research. But when I describe it to other litigators, they nod along in solidarity.
This is the kind of thing we have been thinking about as we develop our legal research and drafting platform. How can we simplify?
Yes, we leverage very good AI—including the generative variety everyone is so excited about. But AI is a means to an end, not the end itself. The end is a simplified legal research and writing process. That's it.
Next week, I think I'll write about another thing I despise about legal research: Putting it all together into an email someone will actually read. We're going to simplify this, too. What else should we fix?