Sunday, July 15, 2018

Q&A with Mike Lissner

For a full background on this issue, please see the following articles:

Why PlainSite No Longer Supports The RECAP Initiative
An Update on the Free Law Project and RECAP
Keeping “Free Law” Free
Invective and Intrigue Within the Free Law Movement Over RECAP Changes
Paul Quintaro, John Fuenchem, and the Double-Edged Sword

The following is an edited version of a conversation I had with Mike Lissner, the Executive Director of the Free Law Project.

PART I: Public Consultation, Public Announcements, and Technical Issues

1. Now that you've answered my questions, I may edit this conversation into a Q&A and publish it. How do you feel about that?

Of course, I figured you'd do something like this, but I'd still hope that you'd consider refraining at least temporarily. We have some blog posts and website updates planned that I think should provide clarity around these issues. Now that we know your points of confusion and of contention, it should be easier to address them on the website.

This may take a few weeks though — we have a few hard deadlines coming up that need to be met and I can't drop those as much as I'd like to. Already they're threatened by the amount of time these emails have consumed. Plus, it's been six months, so another little bit won't change things dramatically. I hope you can be patient if you're willing to wait for these updates. I have a conference in a few weeks that I need to prepare for. After that, this is on deck.

If you won't refrain, I ask that you give me a chance to review your FAQ or blog post before you publish it so that we can discuss any points that are mistaken or otherwise unfair. If you don't do that, then I'll have to do some sort of formal response/correction after you publish, which will just be annoying for all. Better that we resolve such things between ourselves first, if we can.

Maybe if you send me a draft (even if it's just an outline), we can incorporate that into the FLP website so that the information is where people would expect to find it rather than one some other website. What do you think of that?

You can do as you please of course, but since we have these updates planned, I suspect they'll provide much of the information that you're hoping to get out there. Maybe you can contribute to them to make sure they say what you want.

That might be a half-OK resolution to this back and forth.

2. How do you feel about this proposed draft of the Q&A?

Some responses:

1. Why are you bothering? Are you still unhappy about this? Even after I've made the updates to the website? I'm surprised, and I thought we made genuine progress towards the goals you had around transparency and understanding.

2. How are you planning to promote this?

3. I just genuinely don't understand why you're dedicating your time to attacking me and FLP and undermining RECAP. Do you see something better that you'd rather promote or do you have an alternate vision that you think would be better?

I literally sit here every day trying to make RECAP better and make it stronger financially and somehow you're dedicated to attacking me. Why not spend your time doing positive work? Promoting RECAP?

4. If you insist on publishing this, I insist that you add a explanation of the background of our interactions. This should include:

- How many emails you sent to people on the topic of RECAP, and that you sent those emails over a period of six months.

- That I reached out to you to try to satisfy your issues with RECAP so that you would (hopefully) stop lobbying against me, my organization, and RECAP.

- That this was a private email exchange, not an interview or Q&A as you've framed it. To call it a Q&A is an incredibly dishonest misrepresentation.

- That I do not want you publishing this. I don't.

Everything I wrote there is me going to extreme lengths to explain to you why you're doing the wrong thing, that you're damaging my life's work, and that you're doing harm to RECAP. That context is essential and it'd be shockingly dishonest to omit it.

The entire thing is out of context. That's the point I'm trying to get across to you. You took an email exchange and edited it to made it look like a Q&A. That's literally the definition of a different context. As I said before the only fair thing to do here, if you really think you're only providing transparency, is to transparently explain:

1. That I was doing this because you were emailing dozens (hundreds?) of people over the course of six months and showed no signs of stopping.

Perhaps you should have a section in there that says you "refuse to answer" this question about how many people you emailed? Or another section that says you refuse to say who you are?

2. You also, if you care about fairness, should explain that this was a lengthy, private email exchange, not something I was writing for some website you created. Noting the change of context is essential and it must be before the first question for it to be in any way fair.

I strenuously disagree. You were systematically sending false (or at best, misleading) information to our users and supporters and that's why I engaged with you: To help you understand the full story and to find a place where we could all move on with our lives. Without this context, numerous paragraphs in my emails don't make sense, and it's entirely unclear why I'm writing anything to you in the first place.

This is an essential part of the context that should not be left off. Why not include this as an introduction?

No, I don't think it is clear that I reached out to you. That's an important part of the context you're conveniently leaving out, and that belongs at the top if you insist on publishing this.

This is NOT a Q&A. This is you editing my emails into some form that suits you. A Q&A would have both questions and answers that you wrote yourself using the emails as a basis and quoting from them if it made sense to do so. That's very different.

Q: Does FLP ask for contributions from API users?
A: According to the email dated, XYZ, yes/no etc.

That'd be a Q&A. Converting my emails into a Q&A like you've done here is very different. It's literally a different context.

I didn't bother opposing the fact that you were sharing my emails with people or planning to post them — that was off topic at the time, and I didn't think you'd honor such a request — but I do oppose it and would have preferred that you not do so.

Just because I said that you were planning to do this doesn't mean I favor it.

Putting this information in the 19th question means that it's buried. If you insist on doxing my emails and converting them to a different context, this information must go at the top to provide proper context.

3. Why were these changes made with no public consultation or discussion?

Well, there are lots of reasons. One, I just ran out of time to do a better job of this. Writing blog posts takes time and effort, doing community outreach does too, and it was a push to get the extension out the door before the hard deadline from Mozilla. Given how much work it took, I'm happy that we survived their obsolescence of the old extension. I don't think we get nearly enough credit for that, because it wasn't easy. Trust me when I say that I was working so hard my body ached. I just ran out of time to do as much blogging/outreach/whatever as I wanted.

But the other side of the coin is that I have done a lot of outreach and I have attempted a lot of community engagement over the years. You can see evidence of this on lots of places on the site. Bottom line: Engagement on a budget is nearly impossible. To engage a community of users, you have to make serious effort and, ultimately, that takes time (money). Consider this: Can you think of any extension where you'd notice if they did a blog post or sent a tweet trying to engage you? I use a couple, like adblockers and things. I can't imagine I'd know if they did a blog post unless it got mainstream press, but that seems exceedingly unlikely for something like RECAP.

Or another example: When we have new versions of the extension, we often seek beta testers by putting out a tweet or two. If we're lucky, we get one reply. I don't want to say that we'd utterly fail if we made an effort here — who knows, maybe we'd get lucky — but it's a tough thing to do.

And one final reason: We didn't have a whole lot of choice. The old RECAP server was about eight years old. In software terms, that's a few lifetimes. It was impossible to debug it and it had lots of really bad bugs. What's worse, some of those bugs affected some of the most important cases. Other bugs made it so that RECAP would completely stop working for literally days on end until somebody complained and we went and fixed it (which was never easy). The old system had problems. Looking at fixing it, it was super clear that the new system should work differently, and have the Archive integrated directly rather than as a separate thing. So, with that in mind, and with the lessons learned on the old system, we designed the new one. It's MUCH better:

- It saves the HTML that people send us, so if we have bugs, we can fix them even in old content. So far, we've reparsed the RECAP content a few times to fix bugs and extract more content than we ever did before.
- It is fast. Most dockets and PDFs are processed in a matter of milliseconds from when you upload them.
- It creates permalinks almost instantly for sharing breaking cases or documents.
- It provides full text search for everything, including for things like parties and attorneys.
- It allows the kinds of premium services that we can (eventually) use to support it.
- It's debuggable, by which I mean, when we have bugs, we actually fix them now. You may not have noticed, but in any given week, I debug and fix numerous problems in the new system. These are problems that we're noticing now, which have existed for eight or nine years. The old system was so creaky and buggy, we didn't even notice the problems, let alone fix them.

So, summing up: I wanted to do more community outreach. I always do. But whenever we do it, it's not spectacularly successful. On top of that, it's hard and requires a lot of resources. (Have you noticed a trend around FLP needing more resources to do the things you want?) And beyond that, I think the new architecture was necessary to make all the improvements listed above, and so even with the engagement, I'm not sure a different outcome could have happened.

And remember: We had to rewrite the Firefox version of RECAP, and the old RECAP server software was falling apart. It was the right time to redo everything and get a fresh start — which we now have. And this is *very* much for the good for our users.

Look, I don't disagree that community outreach is good, but like I said before, it takes a lot of effort and our experience is that people just don't respond to much, and we just don't have the resources. We're *not* Wikipedia where we have thousands of engaged editors. We're RECAP, we have users, and for most of them, we don't even have their email address.

Look, I get it, you don't like the change and you think we should have done more outreach. Sorry you don't like the change, but it was necessary for technical reasons mentioned above. Just stop using RECAP and move on with your life.

4. I'm asking about the changes in RECAP's distribution model. All of this technobabble is irrelevant.

I strenuously disagree. If you don't understand the technical side of things like RECAP, you can't understand why some of these changes were made. That, *or* you just have to accept that there are technical reasons the changes were necessary and that you don't fully understand them.

OK, then I can address that too, but that isn't what the last question was about. The last question was about why weren't people consulted, which I addressed at length.

As for why the change itself. This is technical, and you already don't want technical explanations, but I'll try anyway. First of all I want to make this very clear because it's been a topic of confusion for you since day one: The RECAP extensions *always* sent the uploads to a RECAP server. It worked in the middle to update Internet Archive, and used Internet Archive as a document storage.

The way that worked was extremely novel, but it was also *extremely* difficult to maintain, resulting in complex bugs that we could never fix and scaling problems that were nearly impossible to resolve. Not being able to scale meant that the more people that used RECAP, the worse the performance got, and that quite frequently an upload from the extension could take days to even have a link in any form. And the bugs, oh the bugs. They were numerous and they were nearly impossible to fix.

So why the change? Because the old way was riddled with problems. This fixes those problems. After nearly ten years. That's something to cheer. RECAP has gotten more improvements and fixes in the past six months than possibly ever before. Think about that.

OK, two things deserve response here. First, I've mentioned this before but you have a critical misunderstanding about how RECAP used to work and about how it currently works. In the past, the extension worked by sending PACER data to the RECAP server, which then used Internet Archive for document and metadata storage. That server was controlled by us, was called the RECAP server, and it was hosted at recapextension.org.

The current version is basically the same, except that:

1. It doesn't use Internet Archive as storage, and

2. It uses CourtListener instead.

The extension itself *never* sent *anything* to Internet Archive. It always sent the content to a RECAP server, which then did processing and then uploaded the content to the Internet Archive.

Does that make sense?

Moving on, you say:

> As Aaron has said, that system was working fine and there was no good reason to interfere with it. And yes, you have repeatedly said there were bugs with the old system, but if you can upload the metadata to CourtListener immediately, there is no reason why you can't upload it to the Internet Archive immediately.

There's a bunch of things that are wrong here. I have gathered that you're not a techie, so I don't know how to get into this in a way you'll believe me, but the bottom line is that Internet Archive only allows document uploads and downloads. I.e., you can push PDFs and HTML to it easily enough (which is why we do that), but if you want to push metadata, you have to upload a file with that metadata.

So, say I want to store the following fact in Internet Archive:

Fact #1: Roses are red.

Great, I put that in a file, and then I upload that file to Internet Archive. Done. But then somebody sends us a file (from PACER, say) that says:

Roses are *pink*

Hm. Is that the same as what we know? Let's check. We:

1. Download Fact #1 from Internet Archive.
2. Compare it to what we just learned.

Hm, a conflict. I guess they can be both pink or red. Let's make a new file that says:

Fact #2: Roses are pink.

And then we upload that to Internet Archive.

OK, so that's fixed. But in the meantime, we had to:

1. Receive the updated fact (from PACER, say)
2. Download the fact we thought we knew.
3. Make a new file with the new fact.
4. Upload the new fact.

That's a *lot* to do every time you get a new fact. And *that* is why the old system was so challenging. All of this might take a minute, or it might take a few days in some cases.

In the new system, by contrast, we have a database we control. Databases are how you normally handle these things. With a database when you receive a fact, you just:

1. Update that fact.

This takes very little processing and it's done in a matter of milliseconds.

Right now, we're averaging more than 100,000 docket entries *per day*, because we've been focused on getting as much content from PACER as possible. That's just too much to upload to Internet Archive in a reasonable way, and it's why we needed a new system.

So you (and Aaron) say:

> that system was working fine and there was no good reason to interfere with it.

Well, that's not right. It was taking days to get a URL for a single upload. It would be clogged up for days processing just one update. It had issues parsing data properly. It was *not*, I repeat *not* working just fine. You might not have known about the problems, but believe me, it was a broken system. To fix it required a new approach, which the new system has.

It was also impossible to fix or improve because it was so old.

It was also impossible to scale to the amount of data we're now getting because it used Internet Archive as a database instead of using...a database.

It was not working just fine. The changes we made were *necessary* to keep it working and to make it scale to the data we're now handling.

You guys are judging a system and saying it was working fine, but you and Aaron had no idea, and how could you?

You also say:

> if you can upload the metadata to CourtListener immediately, there is no reason why you can't upload it to the Internet Archive immediately.

That's also not right. CourtListener is designed specifically to receive content from PACER. It knows what to do with that data, how to handle it properly and make it useful. Internet Archive is more like a dumping ground. You can upload things there, at a given rate, but using it as a database like we were before just isn't a good design — at least not at this scale.

So I repeat: There are *definitely* reasons why we cannot just upload to the Internet Archive immediately, and the old system was *not* fine.

Do you accept this? If not, is there anything I can say that will make you understand that the old system had problems and that those problems required a different type of system? I'm not lying here. I've been the maintainer of RECAP for many years and this is what my experience has taught me.

5. Why didn't you put an alert message in the RECAP browser extension informing users of the proposed changes and seeking their input on them?

Pretty much anything is possible with software. The better questions are, would it work, would it be worth the effort, and do you have the resources to do it. In this case, it might have worked (people might have responded to that kind of a popup). But it wouldn't have been worth the effort (the changes were needed regardless for the technical reasons above). And in any case, we didn't have the resources to do it because we had a hard deadline from Mozilla and only had one person (me) working on everything. Even without doing this kind of thing, the work barely got done on time.

6. Why did you not publicly announce that you were charging competitors money for access, or that sites which refused to pay would only be updated once per quarter, or that CourtListener would have preferential access to court documents?

Really? Because we did announce this change on the blog:

https://free.law/2017/11/13/the-next-version-of-recap-is-now-live/

Quoting:

> As part of our transition to a new infrastructure, we are changing how often we push PACER data to the Internet Archive. Instead of pushing in realtime, as we’ve done historically, the new system starting today will upload PACER data once per quarter.

As for your question, as usual, you have some mistaken understandings and assumptions:

1. For years we have asked that people and organizations contribute to FLP if they are relying on our APIs. That was always true, so it wasn't a change we needed to announce. It's also true, as I've mentioned before, that we do this on a sliding scale and we try to find a reasonable solution for anybody that reaches out to us. The money issue has only been a problem for Plainsite. Not for anybody else I can think of.

2. We did announce that the data would only be pushed once/quarter.

3. We didn't know that Plainsite would react the way they did when the blog post went live, and that they'd entirely refuse to support FLP. In fact, since they had supported us before, it made sense to assume that the API would be a big improvement for them that they'd be willing and even excited to support us again.

4. The "preferential access" as you describe it, was always the case, like I said before. We always had a server that the extension interacted with. It did change that our archive became better integrated with the API server, and the archive became much better, so it made sense to show things in our front end as soon as we got them. But look, here's a snapshot of the RECAP Archive from way back in 2012 (!), before these changes went live:

https://web.archive.org/web/20120107031850/http://archive.recapthelaw.org:80/

In any case, is there a design you'd propose that'd be different? How would it work? We always got the content from the extension first. How could it be any other way and still be a functioning system? There might be ways to do this — maybe — but, again, you don't want to get into the "technobabble", and doing it other ways becomes a real stretch very quickly. The current design makes a LOT of sense technically, and it has numerous benefits for users.

5. If you're saying that RECAP users are in the dark about how Plainsite gets its content, well yeah! They always were, and in fact, they always will be, because most of them have never *heard* of Plainsite, and even if they did, they'd probably not particularly care. The reaction I've heard almost across the board is:

"Wait, Plainsite wants unfettered access to the data so they can run their for-profit business with it? And they aren't willing to contribute *anything*?"

Why *shouldn't* plainsite, and other organizations that rely on our work help support FLP? It's in everybody's best interest.

6. If you're saying that RECAP users are in the dark about when things get pushed to the Internet Archive, I don't know what to tell you. We put it in the blog post, and since the changes went live, surely they noticed that the links in RECAP are going to CourtListener. So...they are surely aware that things have changed in some form, and if they want to dig in more deeply, they can check the blog or email us, or read the code, or figure it out in any number of other ways.

So, once again, I've responded at length to your question explaining why your premise is wrong and why your attack-dog methods are just serving to damage RECAP. I *assume* that's not your goal, so can you tell me the answers to the questions above before I spend any more time on you? What will it take for you to call a truce and move on?

7. Don't you see anything wrong with an ostensibly non-profit organization making anti-competitive changes in secret, with no public consultation, and without announcing the changes to the public?

First, I don't see how you can call these anti-competitive. Do you see how laughable that is? If we were actual competitors with Plainsite and other organizations, do you think they'd have *ever* gotten data from us? Do you think we'd have *ever* given it to them? The fact is, everything needs money to run, and for us a logical place to get it from is for-profit corporations that are the biggest users of the RECAP, as I've explained before. Why aren't you more focused on the fact that Plainsite — a for profit corporation — isn't willing to support the system they rely on at any price? Remember: We tried to make something work, but they didn't want to do anything at any price. If they want to have this data, I'm ready to work with them to find something that works. That's anti-competitive?

As for the "secret" part, give me a break. I've explained in great length how I would have liked to do more outreach on this topic. But, like I said, the priority was to get the system launched according to a deadline set by Mozilla. Would you have preferred that RECAP stopped working and we did more outreach instead? I don't think that's the right balance and there was only time for one or the other.

This whole thing is so nuts. I'm working every day to make RECAP better. To make it continue to *exist*. There would be *no* RECAP for Firefox *at all* right now, if it weren't for the work we do at FLP. Instead, we have a new system that works much better than the old one. RECAP is healthier than it has been since 2010 or so. Have you considered that? Would you have preferred that RECAP for Firefox died or that we didn't get 100,000 docket entries per day?

As for whether people are supportive across the board, I'll say this: Nobody has written up anything about this except for Aaron and Bob Ambrogi. The article Bob wrote was balanced and had thoughts from a variety of people in the industry. Most of those thoughts were supportive or at least understanding.

If, after this, you are still having people respond that they didn't know about the change, doesn't that kind of indicate to you that it's hard to get this message out? Nobody else has taken the bait to write an article. A few people looked into it, spoke to me, and decided not to. Doesn't seem like they agree there's a scandal here.

Look, I don't know what it's going to take to convince you that I'm working hard and doing difficult work for not much money. You'll probably never be convinced. I thought that if I gave you a pound of flesh, and answered your questions, you'd accept that things aren't as scandalous as you think and that you had some mistaken assumptions. But I'm not sure I'm making any progress.

I've already given the past several years of my life to this project, and given you immense quantities of detail about how FLP works, how decisions are made, and why these changes were necessary. What more do you need before you realize it's time to move on?

How do we end this in a way where you move on with your life and I don't have to spend any more time responding to your messages?

Do you accept that:

1. The technical changes were necessary OR that you don't understand the technical part enough to know one way or another?

2. I wanted to do more outreach, but lacked resources to do so?

3. There's no big secret scandal here? FLP is just a small organization that lacks resources to do tons of outreach?

4. The situation with Plainsite is unfortunate, but also it's reasonable to ask them to contribute something as a for-profit organization reliant upon the data?

5. I am accountable in a variety of ways, and I consulted with people before and after making these changes?

6. The new RECAP system is vastly improved and we're fortunate that RECAP for Firefox even continues to exist?

If you haven't yet accepted those things, I'm really not sure what we're doing here. If you have, I don't know why you won't move on.

8. Why did you and the other Board members repeatedly refuse to answer any of my questions? Why is all of this shrouded in secrecy?

I don't feel like it was super secret, but when somebody comes and starts asking lots of accusatory and combative questions, as you did (and continue to do, except, perhaps in your last email), I don't trust them, and I don't (usually) spend my time on them. It's risky for me and the organization, and it's unlikely that you can ever please somebody that's already asking combative questions.

That doesn't make it secret, so much as it makes it...not a good or safe use of my time. Alas, it seems I have no choice but to email with you now even if I think it's not normally the best use of my time. I still find this risky and time consuming, but we're getting there, I think. Maybe you'll be satisfied after this email, who knows?

I can't speak for the other board members, but pretty quickly you were on the offensive. It was (and is) a better use of my time to focus on people that are supporting us rather than ones that are attacking us. If somebody already doesn't like you, it's near impossible to change their mind. And it's rarely worth it.

9. Why are you so reluctant to answer my questions now?

You've absorbed a lot of my time already John, so you need to recognize that any time I spend responding to you is time I don't spend helping the rest of the RECAP users. And I've already spent a lot of time dealing with your emails already.

On top of all this, which I think a reason enough to prefer email, I can't trust you not to use whatever I say against me. You're doing your best to disparage me and my organization, and I'm unsure that you'll ever be satisfied — so far, you haven't been, but I'm trying one last time. If I responded to all your questions in writing, I think there's about a 99% chance there wouldd be a misunderstanding, or that you'd be unsatisfied in some way. And when that happened, you'd have more ammo for your emails.

So I don't think it's a fair way to go forward, for you to ask a gazillion questions, any one of which you can use against me and my organization, if you so choose.

So far, I've spent a lot of time answering your questions, and yet you persist in demanding that I answer more and more. With every question I answer, I take more risk that you'll turn around and use my words against me.

Are you satisfied enough yet to leave me and my organization alone? Do you feel like you've got the big picture or at least enough of it to understand that I'm working hard at this, doing a very difficult thing, and working with extremely limited resources? So far, you've only hurt RECAP via your emails to everybody and now via taking my time. Are you done yet? I sincerely propose we all move on.

We can keep going over the questions, but the answers form a trend. I work hard. I'm not perfect. I could do a better job sometimes, but that's true for all of us. More resources would solve a lot of problems, but that hasn't happened in the many years we've been doing this. The situation with Aaron is pretty bad, but he's unwilling to work with us. We've tried hard to work with him, but it hasn't worked out. Sometimes in life, things don't work out. It sucks, but it happens.

I know you're aching for answers do your array of questions, but I've given you a lot over the past week so you see the big picture. Do you get it yet? Will you move on yet? I don't enjoy this, it's a bad use of my time, and I don't feel you've given me a choice. Do you feel good about that? Are you sure you're fighting the right fight? I want to move on and spend my time doing real work, not satisfying one user that just won't stop sending emails to people.

Just stop using RECAP and move on. Seriously. I think it's time. I think it's WAY past time. I want to work on RECAP, not on satisfying you. Let's do better things with our lives than continue this conversation forever. Do you get it yet?

So. I'll answer your latest question, but I have questions for you before we continue to another of your questions:

1. Will you ever be satisfied? What would this take? I've tried to explain why you're mistaken about a number of things, and to explain why there are subtleties with regards to Plainsite and RECAP that you're unaware of. What more will it take? By now, I think you should realize — or at least begin to realize — that you've made a tempest out of a teapot and that there's really no scandal here. Are you beginning to see that?

2. Do you think it's reasonable to attack an organization like FLP when it's doing good work? Do you think it's reasonable to force me to answer these questions for days and days? You're not just asking idle questions — that'd be fine. No, you're demanding detailed answers to complex questions, and you're forcing those answers under threat of attack.

I think this is totally unreasonable and that you need to stop. Just call a truce and let us move on. I've been begging you for days, but I worry that you are getting pleasure out of this — it seems likely that attacking me and FLP has become a hobby for you. Especially since you've been doing it so long now.

3. Does any part of you think that you're just torturing me for the sake of it, and/or that you're damaging RECAP by doing so? Maybe it's time to stop?

Please let me know if this is beginning to provide clarity into my and FLP's thinking. I'd love to avoid doing this much typing for all 18 of your questions. Big picture, I want you to understand that I sit in this chair for upwards of 40 hours a week for the purpose of getting this data out there. I'm not evil or dishonest like you seem to think I am, but I am trying to make a living so that I can continue doing what I think is important work. Do you think I quit my job and founded FLP to get rich?

Please remember, every email I type to you is real work that's not getting done. The sooner you can be satisfied, the sooner we can all go back to doing good things. Please try to find a way to understand things in a favorable light. You might not be there yet, but please try.

Are we done? Are you done? Can we all move on with our lives? I've replied to your questions honestly and thoroughly.

Once again, I humbly suggest we all find more enjoyable and productive ways to spend our time.

There are now 12,000 words in this email chain. How many will you force me to write before you realize it's time to move on with your life and stop damaging RECAP? Please just move on.

10. You seem to think it's a burden to consult with the community about changes and answer users' questions. Is that correct?

You're wrong. I don't care at all about *your* input. You've been a burden to me and you've done damage to me, my organization, and RECAP. But that doesn't mean I treat everybody this way. I get input from lots of other people and I welcome it. But you, who has been attacking me and combative? I don't want your input. I just want you to leave me and my organization alone.

Everybody I've talked to has told me not to communicate with you at all. Everybody. I'm here today and over the past week in hopes that they're wrong, and that although you have some wrong notions about FLP and about RECAP, I can help you to understand that there's no scandal here. I'm taking a big risk talking to you at all. The more I say, the more chances you have to screw me. But, I'm taking this risk because I'm hopeful you're sensible and that you'll see that I'm one person, working hard for little pay, to do what I think is important work.

Still though, you won't go away. Perhaps I'm wrong and nothing will satisfy you enough for you to end what you see as your righteous crusade to expose RECAP and FLP. Or whatever you think your mission is.

*You* and your questions are a burden on me, yes. Like I said before, I have done lots of community engagement in the past, and I'll do more of it in the future. I believe in building up the RECAP community. But let's be clear here. RECAP is an extension. Extensions don't usually have communities. Besides that, building community and doing outreach takes time. I only have so much of that, so I have to decide, carefully, if it's worth my time to do community outreach or if my time is better spent doing work to enhance the system.

First, nothing you say will convince me that you're not forcing me to write these emails to the detriment of RECAP. What does RECAP get out of these emails? What do I get from it? The only thing I hope to get is you gaining enough of an understanding of me, FLP, and RECAP, that you'll move on with your life.

Second, you can't compare FLP to Wikipedia. Do you know how much time it takes to do what they do? Wikipedia has full time staff doing community work. We don't have full time staff doing any single thing. Like I've said before, I'm not opposed to community outreach, and I've done lots of it over the years for FLP, but it's not easy and it takes time, which is extremely scarce. As with anything I do, it's a trade off. If I do one thing, I don't do another. Right now, I'm writing this email to you. I could be working on appellate support for RECAP instead. I wish I were, but you've given me no choice.

As to accountability and fiefdom and whatever, you, frankly, have no idea who I report to, nor who I consult with. You just don't, because how could you? You assume I report to nobody or maybe just to my board, but you have no idea. I have a board and I consult with lots of other people about changes. I literally consult every day with different people, and we have a slack group where people give their input. Right now, it has something like 50 members that I discuss things with.

I am happy to receive feedback and criticism. I seek and receive it all the time. But, like I've said, I don't care for yours. This isn't an attitude I have towards other people. Only those that have attacked my organization for six months straight.

You know, you didn't have to go and do all those emails just because you didn't like my answers. You could have just...stopped using RECAP and moved on. I'm glad you're passionate about RECAP, but I don't think I'll ever understand why you cared so much and decided to do your emailing campaign.

PART II: Relationships with Other Organizations

11. How much money did you intend to charge PlainSite and other organizations? And why have you been so resistant to revealing the price publicly?

As I've said many times previously, I won't disclose details of agreements that we have with people or organizations. You're just some guy from the Internet (as we all are), and it wouldn't be right for me to disclose this kind of information just because you asked and sent lots of emails. Frankly, the exact details are none of your business, but if you want this info, you can ask Aaron and he can tell you. That's fine by me, but it's not my info to tell.

What I can provide are some details around the generalities that might satisfy you. First, setting a price for access to our APIs is not an exact science. Our goal is to find a sweet spot that's fair to everybody including the researcher or organization and FLP. The general idea here is to have a sliding scale pricing model so that those that can support FLP will, and those that cannot can still get what they need to accomplish their goals. In general, this works very well, and there are lots of people that are using the API. Right now if you check the CourtListener homepage, you'll see that we've had over 500,000 API requests in the past 10 days. That's a big success for our mission!

As applies to Plainsite, what I can tell you is that we didn't ever set a price. I suggested something as a conversation starter for what I thought was fair for a commercial user of the data. Aaron refused to consider any amount. He's running a business using the RECAP data, and it seems fair that he contribute something on an ongoing basis to support the system. (Contrary to what you may believe, it takes a lot to run and maintain RECAP.) Aaron may counter that he has contributed in the past, but those contributions were for specific work that we did, not for ongoing support forever.

Well, non-profits are accountable to the public in a number of ways, but none of those ways involve disclosing business transactions to random members of the public. Non-profits are accountable by providing their tax information, and via their board. I can't imagine you'd do much better demanding contract details from United Way or Goodwill or ACLU. Good luck with that.

12. Why won't you even give a ballpark range for how much you proposed to charge PlainSite even though you already told Bob Ambrogi that it was between $500 and $1,000 a month?

Ah, well, I forgot that was in there. So you've got the answer you're looking for already. Good.

13. Ben Edelman suggests that RECAP could be funded through premium services. Instead of charging other organizations money for access to the RECAP database, did you consider offering any premium services?

Yep. And we even have a few (real time alerts are the big one), and we're working on more to be released soon. I think they'll be an important part of our blend of funding along with donations, grants, and client work.

But...this is a tricky thing to do as a non-profit, because you can only sell things like this if you can demonstrate that they're in line with your mission. If you sell too much stuff like this, the IRS (rightly) starts looking at you and thinking "Hey, they really kinda look like a for-profit business." So you have to be careful not to do this too much unless the premium features are part of your mission. But we have room in our finances to do more of this, and we are working on it.

What I think would really be powerful would be to keep track of how much money people are saving with the extensions, then ask for some percentage of that every so often or on an ongoing basis. E.g., "Automatically donate X% of my PACER savings to RECAP." I think a lot of people would go for that, but it's tricky due to privacy issues (we don't track RECAP users), and it's hard to implement generally. But I think it'd work — maybe we'll get there one day. That gets at another problem: Building these kinds of things takes time (money), and so there's always a chicken and egg problem. But we're working on that.

14. What's the nature of the micropayment auction system you are/were planning to implement?

We're not planning to implement something like that, so much as thinking about how it would be a useful and interesting feature. The big picture: RECAP is saving people money. What if they could donate some percentage of that back to FLP as support for RECAP? The more value you got from RECAP, the more you'd donate. It could work nicely, but it's probably riddled with privacy challenges like so many of these things.

15. Did you consider any less drastic measures besides denying access to other organizations if they refused to pay?

Well, as I said already, when we spoke to Aaron and discussed money, I only suggested an amount, and at that point he refused to contribute anything, and then took this nuclear via his blog. If he had said the number was too high, or that he didn't make much money with Plainsite, we would have found a way for it to work.

Our goal is to get this data to as many people as possible. Part of that means we need to charge somebody for something so that FLP and RECAP can continue existing. If FLP collapses, RECAP will too. Money makes the world go 'round, but despite running a for-profit business on top of the data we provide, he wasn't willing to make ongoing contributions to support us. I find that really disappointing, and I was surprised by the stance he took.

As for other ways to make money, we've tried and considered most of them over the years. We never take changes lightly and we try to proceed using the experience we have and considering our mission as the most important thing. So, you can charge for the extension, but then nobody would use or install it. You can ad advertising to your website, but we tried that and it didn't bring much money in (at the same time, the ads were horrible). You can do fundraisers. We do, and they help, but they're not enough. Or you can do the obvious thing, and build useful services for people that have the biggest demands for data, and you can work with them to find a price that makes everybody successful. It should be in their interest to to provide ongoing support if they rely on your service, and it shouldn't be hard to find a reasonable price point on a sliding scale that works for everybody. Have you considered there may be a lot of people happily using and supporting the RECAP APIs that you don't know about?

Yes, we considered many things. We tried many things. And this is what we're trying now.

But ultimately, your premise is a bit off. We tried hard to work with Aaron, but he was unwilling to contribute anything to support his main source of data for his for-profit business. I don't see this as us refusing him data. I see it as him refusing to support the thing that he has relied on for something like ten years. We didn't ask for much. In fact, we didn't even set a price. He refused to consider supporting us at any ongoing amount.

16. Did you give any consideration to the impact these changes would have on the users of PlainSite and other sites which refused to pay?

As I've said before, we tried to find a way to work with Aaron, but he was unwilling to really even discuss the topic of ongoing contributions. So I don't see it so much as us denying him the documents as I do him refusing to contribute to something he relies on. Maybe it's both. But regardless, yes, I've considered how that would affect Aaron and his users. I'm not happy about this situation, as I hope I've made clear numerous times.

17. Besides contributing to your salary, is there a compromise, such as an in-kind donation, that would lead you to grant PlainSite real-time access to the documents?

There's probably a way forward, but I don't think you're going to broker it. To broker such a deal would require that I trust you, and I don't. I don't even know your name, and you don't trust me to tell it to me. I think a deal could involve all of these things you mention, yep, but I don't think Aaron is interested.

As for the past contributions he's made, I'm glad you brought these up. All of these were to build specific features for RECAP. That kind of thing is good for RECAP feature-wise, but it doesn't go towards ongoing expenses or ongoing maintenance. In other words, these were essentially work-for-hire types of contributions. He gave this and in return he got that.

I also find it weird how he always includes the funds from Princeton in his accounting, but that's another topic. There was also some weirdness about how even those funds worked out that made a lot of people frustrated. This was before my time, so I don't know what that's about, but point being that nothing is as simple as it seems.

In any case, his last contribution was in 2014. That was for a concrete piece of work he wanted, and it was a long time ago for something that his (for profit!) business relies on every day.

18. How do you feel knowing that the United States Courts Archive has shut down due to the changes?

As for United States Courts Archive, you've got your facts all wrong. I talked to him around the time of the changes, and he effectively said, "Oh, so you're putting everything in RECAP online? Then my work is done." If you go to the United States Courts Archive website, you'll see he makes no mention of being upset, and instead links to RECAP and CourtListener four times. The guy who ran that site didn't put any ongoing work into it, and had a different full time job. When he learned we were putting everything online, he seemed fine with it in a "mission accomplished" kind of way. Things with him really never got as far as money.

I'd be surprised if the guy running USCA talked to Aaron. I think Aaron is just using the same assumptions that you are on this topic. I did talk to the guy running it to get his perspective, and you have to admit, if he was angry, his homepage wouldn't look the way it does. He'd say if he was upset, or he'd at least not drive all his traffic to us. Right?

19. Actually, he did speak with Aaron.

Interesting that they spoke. I guess it's possible he told me and Aaron different versions of the truth about why he shut it down, or that he had a variety of reasons, or that Aaron and I focused on different parts of it. He seemed really not very put out though when I talked to him, I'll say that. And he redirects his traffic, which I think is a pretty big indicator. And we never got to the topic of money, so it wasn't that either...

20. If there's no scandal here, why did the operator of the United States Courts Archive and several other individuals refuse to discuss the matter on the record with Bob Ambrogi? Ambrogi told me that trying to get information was "like investigating Watergate." Why is this shrouded in secrecy?

Yeah, can you blame [the operator of the United States Courts Archive] for not wanting to get involved?

I think people that got your letter probably decided they didn't want to get into it, or didn't care enough to go on the record. I don't know, but I don't think that indicates that there's some sort of scandal. Some people — some very important ones in this world — did go on the record, and they didn't seem outraged to me.

I've also discussed this with many others in the community both before the change, and as a result of your letters. Their response has been unanimously supportive and understanding, with the two exceptions you know: Ben Edelman and Aaron.

I think people just don't want to get involved. I don't know.

PART III: Money, the Free Law Project, and Donors

21. Why did you collect a salary of $90,010 in 2016 even though the FLP's total revenues were only $54,528? And if you actually made less than $90,010 as you claimed in Bob Ambrogi's article, why does the Form 990 say $90,010?

I sent a correction to Bob for that, since it was off the cuff during a phone interview. I wrote:

> First, the 90k salary figure is correct — for 2016. In 2015 we were more focused on getting things rolling than on getting me paid properly, so 2016 adjusted for that. Last year was a normal year at about 62k, so I think the important number is that from 2015 to 2017 the average was just under 60k — pretty meager, honestly, for a CTO, Exec Director or Developer in the bay area, and much less than I used to get before I started FLP.

Bob said he'd post this, but I don't think he ever did. In any case, the answer to your question is: Your facts are wrong/incomplete; dig deeper. Also, please don't miss the fact that I was doing this and not collecting a salary for much of 2015. We were so focused on doing this work that we didn't figure out accounting. It just wasn't as important as doing real work. And, sure enough, that's what you've been using as part of your ammo in your emails. I guess a good deed never goes unpunished.

I've been doing this full time for something like four years I think. Go back and see how much money I've been paid over that period, and consider how much somebody with my skills could make doing a different job in the bay area. $60k is probably less than half of what I'd make if I took a tech job. That's a HUGE gap that we're trying to address.

22. Ben Edelman suggests that RECAP's expenses shouldn't be particularly high, especially considering that open source software development often does not entail paying developers anything at all. How do you reconcile this with a salary of $90,010?

Almost everything goes towards my salary. This is pretty common in small non-profits with low outside expenses. We do everything we can to keep our expenses extremely low outside of having to pay me.

We've been open source for nearly ten years. We've yet to see the crowds banging down our door to fix bugs and develop useful features. We have a few volunteers, and they're awesome, but what Ben's hoping for here is a free lunch. This is a nice idea and it's the dream of open source since forever, but that only exists in very limited arenas.

An example. The Linux kernel is probably considered the most successful open source project ever. They, helpfully, have a foundation that does an report every so often. Here's the 2017 report:

https://go.pardot.com/l/6342/2017-10-24/3xr3f2/6342/188781/Publication_LinuxKernelReport_2017.pdf

Look on page 15, and you'll see:

> well over 85 percent of all kernel development is demonstrably done by developers who are being paid for their work.

(emphasis mine)

And:

> ...the volume of contributions from unpaid developers [...] is 8.2 percent

That's for the Linux kernel.

I wish Ben was right, and we had a stampede of developers banging down our door to work on RECAP. We just don't, and trust me when I say that volunteers are very limited in the kinds of things they are willing and able to work on. They can do bug fixes, but they pretty much always prefer to build new features. Features that, in turn...would have more bugs nobody would want to fix....

Having me to work on RECAP and CourtListener full time, for years, is important. That takes money.

23. Your organization's name is the "Free Law Project," and you constantly argue, on your website and in court filings, that court documents should be free and freely available to everyone. Is one person's salary worth compromising this principle?

Look, I've spent a lot of time discussing things with you. If you believe that RECAP would be just fine without anybody working on it, fine. But what do you think I do here? I work on RECAP pretty much every day to keep it running, add new features, expand and improve how it works, etc.

Right now, for example, there'd be no RECAP for Firefox at all if I didn't sit here and work on things. There'd barely be a RECAP server. To the people that think RECAP can just exist without somebody working on it, I disagree. Why? Because I've poured hundreds of hours into making it work and I know that work was important to its continuing (and improving) functionality.

If you have any doubt about the need for somebody doing this work, have a look at our (totally open; totally transparent) bug backlog. It's long and detailed and it's full of requests and bugs from users that want RECAP to be better in one way or another. Right now we have something like 300 open issues across CourtListener, RECAP, and Juriscraper. Not all of these are for RECAP, but a lot are, and the others are important too.

An example: Today somebody reached out with thousands of PDFs they want included in the RECAP Archive. Would you prefer that I add those to the archive (for the good of everybody) or that nobody respond to his message at all? That work won't get done for free.

We do provide free access to legal materials to the vast majority of people and organizations. You can go to our website, you can download bulk data, we push thousands of PDFs to Internet Archive nightly, and most of our APIs are in fact free — they have been for years. For some of our APIs — the ones that require the most work — we ask for contributions from those that can afford them. We don't set an amount, and we do so on a sliding scale.

I honestly don't understand your vision for Free Law Project or RECAP. Do you think I should quit, get a new job, and then we all pray that everything keeps working smoothly? It doesn't work like that. There's work to be done. That work requires staffing. Staffing requires money.

I think those questions are answered. The old system was near death and it costs money to fix, maintain, and expand it. That money has to come from somewhere. Operating costs include having somebody (like me) that can jump in and fix bugs, reboot servers, all that stuff.

Maybe I haven't driven this home enough, but Mozilla deprecated the system that the old Firefox APIs used. If it wasn't for me working for months, there would *be* no Firefox extension right now and the server would be barely hanging on. Is that what you prefer? Or do you think those things should be somehow done by free labor?

This isn't a deposition. If you're not prepared to make some assumptions in good faith based on what you know at this point, I don't know how to continue here. The bottom line is that running, maintaining, and enhancing RECAP takes time. The old system was not fine. It needed a lot of work. Work takes staffing. Staffing takes money.

If it wasn't for FLP paying my salary, there would be no RECAP for Firefox right now and the server would still be in very bad shape. I don't understand how you can't make the connection that me getting paid for this work leads directly to RECAP continuing to exist and to thrive, and that that is how information is available at all. You're caught up on the perfect, and unwilling to accept the good.

It's insane to me that I've dedicated nearly a decade of my life to this work and that you're coming after me for wanting compensation for it. Do you see other people doing this for free somewhere? This is mind-boggling.

24. $62,000 a year is not "free labour." Why is a $6,000 increase (or any increase) in your salary necessary for RECAP to exist?

Once again, we never settled on an amount, though I'm not sure that's here nor there.

As for the rest of my salary, you can debate it if you want. $60k is less than half of what I could get doing this same work elsewhere, and if you want to have good people doing a job, you have to pay competitive wages. The board supports this amount of money (both past board members and current ones), and it's not an unusually high amount in the Bay Area, where we are based.

25. How much of the FLP's revenues are spent on buying documents from PACER?

[refused to answer]

26. How much of the FLP's revenues are spent on the servers and bandwidth?

[refused to answer]

27. Why are the minutes of your Board meetings not posted online? Will you release the minutes of the Board meeting(s) where you decided to make the RECAP changes?

Sorry, we don't release the minutes currently. We probably could, but I don't want to go through them to see what is/isn't in there. Probably nothing, but we do talk about people we'd want on the board, and things of that sort. Some of that gets personal.

But mostly they're pretty mundane bullet-point kinds of notes. I'll consider posting them in the future. It's not a bad idea, but again, we're not Wikipedia.

28. What changes have you made to address concerns about your lack of transparency?

I'm still at this conference, but I managed to find a minute to make some website changes. I think these address a lot of your concerns if not all of them, but please let me know if they raise any others.

The changes appear on the FAQ section of the RECAP site, where there are a couple new Q&A's:

https://free.law/recap/faq/#recap-and-internet%C2%A0archive

And at the entry point to the APIs, where we've spelled out our process and expectations more clearly:

https://www.courtlistener.com/api/

You mentioned that Aaron thought that we should be asking more of law firms, so I made sure that in both of these places, we mention them. That was a good call. I also went so far as to provide actual dollar figures at least in the RECAP FAQ's, like you suggested.

There are a handful of other smaller tweaks throughout the system to clarify language. You can find those in the source code, if you're interested, but they're not terribly thrilling.

I don't think at this point we'll be doing any blog posts specifically about this topic. To my mind, the above changes make things pretty clear.

What do you say, can we put this whole thing to bed?

29. On your updated FAQ, you say "we hope you’ll consider contributing to support Free Law Project in exchange for your usage of the APIs. We only ask that you contribute..." Is payment mandatory or voluntary? And why is the wording so ambiguous?

[refused to answer]

30. Why do you use a "pay what you can" model instead of the "pay what you want" model which Wikipedia uses?

[refused to answer]

31. Your top three donors in 2017 contributed $81,091, which was 85% of your revenue for that year. Why are the names of those donors omitted from your 2017 Form 990?

Disclosing top donors isn't required in your public 990, and like many non-profits, we don't disclose ours. I would normally disclose this to somebody that asked, but since you're going out and sending the kinds of emails you are, I don't want to give you anything you could use as ammunition. Sorry, but why would I tell you this?

32. Are the top three donors individuals, for-profit corporations, or other non-profit organizations?

[refused to answer]

33. Of the $95,459 in contributions you received in 2017, only $18,792 of it was "gifts," while the other $76,667 was "program service revenue." What are your top three donors receiving in return for their contributions? Are they simply getting access to the RECAP database or are they "investing" in the FLP?

[refused to answer]

34. If you are a non-profit organization, why does 80% of your funding come from "program service revenue?" Are you actually running a for-profit business?

[refused to answer]

PART IV: Personal Questions from John Fuenchem

35. Why did you falsely accuse me of being a pseudonym for Aaron Greenspan?

I couldn't, and still can't imagine anybody caring as much as you do about such an esoteric thing without having a financial incentive like Aaron does. You being Aaron was the most likely explanation, and I still have no reason to believe that you're not Aaron's silent partner or something like that, quietly using these emails to build a case against me. I'll never know what animates you, and that's fine with me.

On top of this:

- You don't exist in Google. This is nearly impossible for most people.
- Nobody knows you, but you seem to know everybody. (Sounds like an alias.)
- You wouldn't reveal your identity to Bob Ambrogi.
- You have a Canadian Yahoo! account. Why would a Canadian care about PACER, much less RECAP?
- You won't have a phone call with me.
- etc.

There are lots of reasons to think you either *are* Aaron or are somehow closely affiliated with him. I don't much care at this point, provided you go away, but are you ready to provide proof of your identity now? Seems like a simple request. If not, I'll continue to assume the above, but I'll never really know. Honestly, other people care a lot more than I do. I just want to stop seeing your name in my inbox.

36. Why have you made false allegations of "harassment" and threats of legal action against me? Do you disavow the accusations and threats Brad Johnson made against me?

I spent time talking about you because when somebody is attacking your organization and emailing your supporters — for six months! — you end up trying to figure out how to take care of it, and you wind up talking a lot about that person. My hope is that by emailing you now, I'll never have to think about you or talk about you again. That's my only goal — to convince you to leave me and my organization alone.

As far as I'm concerned, you *have* harassed me. You have attacked my organization with mistaken assumptions. You have forced me to write thousands of words explaining various things. You have spent six months ferreting out people's email addresses, and trying to hunt down my allies and supporters. You've sought out local and national journalists, and wasted a monumental amount of my time. If that's not harassment, I don't know what is.

As for legal action, if this thread ends peacefully, I plan to focus on RECAP. Like always.

Brad doesn't speak for me or FLP.

37. Will you withdraw your allegations and threats against me, and apologize for them, publicly or privately?

I don't think I care about apologies, frankly. I think you owe me one for all of the things I've outlined in these emails. All of the time you've spent sending wrong information and damaging me and my organization. The stress you've caused.

I think you've got this totally backwards and that your righteous crusade was a huge mistake. I think it will take you many years to realize that, but I hope that you eventually will.

Was I wrong about you being Aaron? Time will tell. If so, I apologize for that, and I'll be happy to share that news publicly. But at the same time, I'll never know your association with him, and I don't think it was a wild assumption. It's the one everybody is making.

That said, I don't much care about apologies. I just want to stop seeing your name in my inbox. We don't have to be friends for this to end.

I don't know when I [accused you of being a pseudonym for Aaron], but I went ahead and apologized publicly anyway, in a write up I just posted on Twitter that summarizes some of our exchange:

https://twitter.com/mlissner/status/1003502299291873281

If anything in there bothers you, please let me know. I did my best to be completely fair and accurate so people can have an update on this.

Posted by at No comments:
Subscribe to: Posts (Atom)