WEBVTT - This Is What Happens When Governments Build Software 0:00:10.119 --> 0:00:14.120 Hello, and welcome to another episode of the Odd Lots Podcast. 0:00:14.200 --> 0:00:16.560 I'm Jill Wisenthal and I'm Tracy Alloway. 0:00:16.760 --> 0:00:19.120 Tracy, you know, what I felt was a really interesting 0:00:19.200 --> 0:00:21.720 thing in a recent debt ceiling episode that we did 0:00:21.920 --> 0:00:24.439 was about all the coding it would take if the 0:00:24.480 --> 0:00:27.680 Treasury were to introduce a new kind of a bill. 0:00:27.960 --> 0:00:31.280 Yes, that was interesting to me too, because I think 0:00:31.320 --> 0:00:34.639 there's a perception out there that if Treasury wakes up 0:00:34.720 --> 0:00:37.680 tomorrow and decides to sell a new type of bond, 0:00:38.280 --> 0:00:40.320 they could just do it. But actually there are all 0:00:40.360 --> 0:00:43.160 these back end changes that would need to be made. 0:00:43.720 --> 0:00:46.879 And as you and I know from multiple conversations at 0:00:46.880 --> 0:00:50.760 this point, it feels like government technology can be a 0:00:50.760 --> 0:00:53.760 little bit clunky sometimes. Is that a fair way of 0:00:53.760 --> 0:00:54.320 putting it? 0:00:54.320 --> 0:00:54.640 It is? 0:00:54.760 --> 0:00:57.000 I would say, you know, in defense, I mean get 0:00:57.000 --> 0:01:00.600 into yes, you know, corporate technology can be clunky. But 0:01:00.760 --> 0:01:04.640 that point reminded me that, like we've done, software is 0:01:04.640 --> 0:01:07.440 like a fascinating thing. And we talk a lot on 0:01:07.480 --> 0:01:10.200 the show. I think about like the sort of built economy, 0:01:10.319 --> 0:01:13.400 like the Chips Act and battery investment and semiconductor and 0:01:13.440 --> 0:01:16.440 all these things are like can the US build things again? 0:01:16.520 --> 0:01:20.280 Like as a government state capacity these ideas, but in 0:01:20.319 --> 0:01:23.759 the twenty first century also like building software in tech 0:01:23.959 --> 0:01:24.920 has to be part of that. 0:01:24.959 --> 0:01:27.800 Question, right, So it's one thing to announce that you're 0:01:27.840 --> 0:01:31.240 going to do this new thing, but almost everything that 0:01:31.280 --> 0:01:34.800 we do nowadays comes with some sort of software requirement. 0:01:35.120 --> 0:01:38.520 Like I'm sure there is a system for processing applications 0:01:38.560 --> 0:01:41.479 for chips, funding and things like that, and someone has 0:01:41.520 --> 0:01:44.600 to actually build that. So there's two questions contained in 0:01:44.640 --> 0:01:48.040 the can America build things question? It's like, can it 0:01:48.120 --> 0:01:50.800 build the actual things and then can it build the 0:01:50.840 --> 0:01:53.200 software systems that would allow it to build the things? 0:01:53.320 --> 0:01:56.280 And you know, in the pandemic period, we got sort 0:01:56.320 --> 0:01:58.200 of smacked in the face, I would say, by this 0:01:58.320 --> 0:02:02.120 realization that the saw fair component is not trivial at all. 0:02:02.200 --> 0:02:05.680 And most notably it was like every state had some 0:02:05.880 --> 0:02:09.440 sort of problem with their unemployment insurance system and there 0:02:09.440 --> 0:02:12.200 were just such a surge in claims obviously in those 0:02:12.200 --> 0:02:14.160 initial you know, really for that first year, just on 0:02:14.160 --> 0:02:16.920 the scale that none had ever seen. And it reminded 0:02:16.960 --> 0:02:20.480 people's like, oh, yeah, fifty different stage systems much with 0:02:20.600 --> 0:02:23.480 like tech that was probably built by some contractor, and 0:02:23.720 --> 0:02:25.600 I don't know the eighties or nineties and the people 0:02:25.639 --> 0:02:28.280 who knew how to run it have left, and you know, 0:02:28.280 --> 0:02:30.520 it was a I gues Eventually, I guess it got 0:02:30.760 --> 0:02:34.200 smoothed out, but it sort of exposed like, Okay, why 0:02:34.200 --> 0:02:36.399 did that happen? And what else is lurking out there 0:02:36.400 --> 0:02:38.400 that we don't really know how it works until it's broken. 0:02:38.560 --> 0:02:39.919 I'm sure this is going to end up being a 0:02:39.960 --> 0:02:42.920 classic episode where we mentioned Kobyl quite a few times, 0:02:43.000 --> 0:02:45.880 like it's inevitable, right, it's coming, you can feel it. 0:02:45.919 --> 0:02:47.640 One day, We're going to do it like actually, just 0:02:47.680 --> 0:02:50.840 like what is Kobol episode where we just like focus 0:02:50.880 --> 0:02:52.720 on that directly. But yes, we're going to talk about 0:02:52.760 --> 0:02:54.280 software and we're going to talk about it from the 0:02:54.320 --> 0:02:57.480 public sector perspective. What does it mean when the government 0:02:57.560 --> 0:03:00.480 tries to fix something, what happens when the govern tries 0:03:00.480 --> 0:03:02.880 to build something, what happens when the government tries to 0:03:02.880 --> 0:03:05.799 buy something. Some of these themes we talked about them 0:03:05.800 --> 0:03:10.240 from a private sector perspective with Patrick McKenzie earlier in 0:03:10.280 --> 0:03:12.160 the year, but more to do on this topic. 0:03:12.480 --> 0:03:14.800 Yeah, I'm excited. I have a lot This is going 0:03:14.880 --> 0:03:17.880 to be a really good chance to actually ask like 0:03:18.320 --> 0:03:22.280 the decisions that are going into some of these things 0:03:22.320 --> 0:03:25.600 being built. I'm still floored every time I log into 0:03:25.680 --> 0:03:28.000 the Treasury Direct website that you know you have to 0:03:28.000 --> 0:03:31.959 click the little the buttons like the actual letters and 0:03:32.040 --> 0:03:34.640 numbers to type your password. It won't let you just 0:03:34.720 --> 0:03:35.640 type your password. 0:03:36.240 --> 0:03:37.680 How did that come to be? 0:03:38.200 --> 0:03:39.279 I have questions? 0:03:39.880 --> 0:03:42.720 That is a great question. So we have I think 0:03:42.840 --> 0:03:46.720 literally the two perfect guests to discuss this, two guests 0:03:46.720 --> 0:03:51.280 with extensive experience answering and working on exactly these problems. 0:03:51.280 --> 0:03:53.880 We're going to be speaking with Jennifer Palka. She is 0:03:53.920 --> 0:03:57.200 the founder of Code for America. She helped found the 0:03:57.320 --> 0:04:00.880 US Digital Service. She co chaired the cal Fornia task 0:04:00.960 --> 0:04:05.480 force on fixing its unemployment insurance system during the pandemic, 0:04:05.600 --> 0:04:08.680 and she's the author of a new book called Recoding America, 0:04:09.080 --> 0:04:11.240 and we're going to be speaking with it. Dave Guarino, 0:04:11.360 --> 0:04:15.200 who's a number of things currently an independent consultant researcher 0:04:15.240 --> 0:04:18.520 focused on this stuff. Founding engineer at Code for America, 0:04:18.839 --> 0:04:22.920 helped create get cal Fresh, which was like allowed Californians 0:04:22.920 --> 0:04:28.440 to easily access snap has recently worked on the unemployment 0:04:28.440 --> 0:04:32.279 insurance modernization at the Department of Labor. Also worked at California, 0:04:32.360 --> 0:04:35.880 also worked in California. So two people's extensive experience on 0:04:35.960 --> 0:04:38.839 exactly these things. So, Jennifer and David, thank you so 0:04:38.920 --> 0:04:40.000 much for coming on odd. 0:04:39.800 --> 0:04:41.720 Lots great, thanks for having us. 0:04:41.880 --> 0:04:43.279 Absolutely great to be here. 0:04:44.440 --> 0:04:46.640 Now I want to ask Tracy, could I steal your 0:04:46.640 --> 0:04:47.160 first question? 0:04:47.200 --> 0:04:47.360 Yeah? 0:04:47.360 --> 0:04:47.720 I do it. 0:04:47.839 --> 0:04:48.920 How has just happened that. 0:04:48.880 --> 0:04:52.279 A government website doesn't let you type in something and 0:04:52.320 --> 0:04:55.479 you have to like click buttons like that like resemble 0:04:55.600 --> 0:04:57.960 mini keyboard on a website. What happened when you see that? 0:04:58.320 --> 0:04:59.880 Do you have like in your mind you're like, oh, 0:04:59.920 --> 0:05:02.520 I know exactly the meeting that caused this to happen. 0:05:02.720 --> 0:05:05.400 I think Dave has been in those meetings and has 0:05:05.440 --> 0:05:07.919 come out like with his hair on fire, fuming, So 0:05:08.040 --> 0:05:09.760 I will let him answer that. 0:05:09.360 --> 0:05:12.120 I mean, well, I don't have Yeah, I wasn't in 0:05:12.160 --> 0:05:15.880 that specific meeting. Maybe the best answer I can give is, 0:05:16.400 --> 0:05:20.719 and there's deeper issues behind it, is that the way 0:05:20.839 --> 0:05:23.359 Garment tends to build technology. The reason for that is 0:05:23.400 --> 0:05:27.480 no one wrote down specifically as a requirement upfront that 0:05:27.600 --> 0:05:30.800 people should be able to use the keypad and the 0:05:30.880 --> 0:05:35.680 numbers on their keyboard to enter the digits, and so 0:05:35.720 --> 0:05:37.920 someone's like, well, I've met the requirements. Like the requirements 0:05:37.960 --> 0:05:41.120 are that someone can do this, they can enter it, 0:05:41.160 --> 0:05:43.680 and the way they do that is they click in 0:05:43.760 --> 0:05:46.359 this somewhat I suppose insane way. 0:05:46.640 --> 0:05:49.160 This is gonna be a conversation or like bang your 0:05:49.160 --> 0:05:51.120 head on a disk a bunch. I feel like as 0:05:51.160 --> 0:05:52.880 we listen these answers, anyway. 0:05:52.680 --> 0:05:53.400 Yeah, I get ready. 0:05:53.480 --> 0:05:56.480 Okay, Well, let me ask maybe a big picture question 0:05:56.760 --> 0:06:00.560 to start, but what is the process by which government 0:06:00.640 --> 0:06:04.800 software is developed? Because you know that answer just then 0:06:04.960 --> 0:06:07.560 it makes it sound like someone gets a mandate we 0:06:07.640 --> 0:06:10.640 need a website, for instance, through which people can buy 0:06:10.920 --> 0:06:13.960 government bonds, and then someone else goes off and I 0:06:13.960 --> 0:06:16.600 guess they execute it along the lines of the mandate 0:06:16.640 --> 0:06:17.280 that's been set. 0:06:17.360 --> 0:06:19.000 I mean, I'll jump in on that one. I mean, 0:06:19.160 --> 0:06:21.320 first of all, it is changing. So just before we 0:06:21.400 --> 0:06:24.200 get everyone sort of bounding their heads on the table, 0:06:24.240 --> 0:06:26.800 there's a lot of movement in good directions. But what 0:06:27.000 --> 0:06:29.880 has been the process is exactly that there's sort of 0:06:29.960 --> 0:06:33.600 this mandated often comes from legislation or regulation that very 0:06:33.600 --> 0:06:36.120 frequently now like says there will be a website that'll 0:06:36.120 --> 0:06:39.440 even put the URL in the legislation, and then it 0:06:39.560 --> 0:06:43.440 kicks off what's a really long process that does exactly 0:06:43.600 --> 0:06:48.320 what Dave was saying. It is first centered around requirements gathering, 0:06:48.480 --> 0:06:51.599 and that process of just requirements gathering can easily take 0:06:51.640 --> 0:06:53.920 ten years. Now, when it's something like healthcare doc of 0:06:54.320 --> 0:06:56.400 you know, and they have a launch date that's three 0:06:56.440 --> 0:07:00.320 years out, they don't take ten years, but they still 0:07:00.320 --> 0:07:04.520 gather sort of every imaginable requirement and they kind of 0:07:04.560 --> 0:07:08.720 throw it in one giant bucket. And that's how the 0:07:08.920 --> 0:07:12.240 RFP is created. The request for proposal, that's what vendors 0:07:12.400 --> 0:07:16.320 bid on, and it's sort of like an undifferentiated, unprioritized 0:07:16.400 --> 0:07:20.000 mess of everything everyone could think of, plus things that 0:07:20.040 --> 0:07:22.960 have sort of pulled over from previous RFPs, plus a 0:07:23.000 --> 0:07:28.000 bunch of compliance stuff around security, compliance stuff around well 0:07:28.080 --> 0:07:30.720 various other things. Dave can maybe chime in, but what 0:07:30.760 --> 0:07:34.080 that isn't is product management. Right, there's no one saying 0:07:34.200 --> 0:07:37.600 here's the important things this should do. And as Dave said, 0:07:37.640 --> 0:07:40.560 there's never a requirement for it to just work. So 0:07:41.040 --> 0:07:43.600 then the vendor, I mean, I'll skip several steps in 0:07:43.640 --> 0:07:45.720 the middle where you know, it takes a long time 0:07:45.760 --> 0:07:47.960 to bid it out, and then there's like a protest 0:07:48.080 --> 0:07:51.000 because one vendor thinks the other vendor shouldn't have got it, 0:07:51.040 --> 0:07:52.600 and that delays it a couple more years. But then 0:07:52.640 --> 0:07:55.960 they go into development and their job, as Dave said, 0:07:56.040 --> 0:07:58.760 is just like check all these boxes, and it's thousands. 0:07:58.800 --> 0:08:02.480 I mean, when we were at California, State of California 0:08:02.560 --> 0:08:05.679 working on unemployment insurance, they were actually about to put 0:08:05.720 --> 0:08:10.240 out a Business System Modernization RFP. Actually I think about 0:08:10.240 --> 0:08:12.320 it to award it to a vendor, and I think 0:08:12.320 --> 0:08:15.800 it had seven six seven hundred requirements. We can check that, 0:08:15.840 --> 0:08:19.320 but it's in that's really a normal number of requirements. 0:08:19.360 --> 0:08:21.880 And so they do all these things that where you 0:08:21.880 --> 0:08:24.240 can check a box, but what they don't do is 0:08:24.280 --> 0:08:27.560 actually check that it works. So I mean, maybe i'll 0:08:27.600 --> 0:08:29.480 just tell a quick story from the book. And you know, 0:08:29.520 --> 0:08:34.480 there was famously this application for veterans' healthcare benefits at 0:08:34.520 --> 0:08:39.200 the VA that didn't work outside the building and they 0:08:39.840 --> 0:08:43.240 didn't couldn't see it. So basically, somewhere in the specs 0:08:43.360 --> 0:08:46.640 it had that this form needed to work on a 0:08:46.800 --> 0:08:51.840 very specific and outdated combination of Internet Explorer and Adobe 0:08:51.880 --> 0:08:54.600 Reader and The reason no one knew that it didn't 0:08:54.600 --> 0:08:57.000 work inside the building is that's how all the computers 0:08:57.040 --> 0:08:59.439 in there were set up. But if you were outside 0:08:59.440 --> 0:09:02.800 the building, had any other possible combination of those two 0:09:02.800 --> 0:09:05.360 pieces of software, it literally wouldn't load. And so they 0:09:05.360 --> 0:09:09.520 had very very few people applying for these benefits online, 0:09:09.920 --> 0:09:12.920 and you know, veterans were really, really, really frustrated. And 0:09:12.920 --> 0:09:16.720 it took a team going out there recording a veteran 0:09:16.880 --> 0:09:19.679 who had tried to do this dozens and dozens of times, 0:09:20.120 --> 0:09:24.360 bringing that video back and showing it to the Deputy 0:09:24.400 --> 0:09:26.560 secretary for them to be able to say, oh, okay, 0:09:26.600 --> 0:09:30.000 actually there is something to be fixed here. Okay, you know, 0:09:30.040 --> 0:09:31.720 you can go ahead and make a new form. But 0:09:31.800 --> 0:09:34.760 up to them, they said, sorry, it's fine, we're looking 0:09:34.760 --> 0:09:37.719 at the requirements. The requirements have been met. There's technically 0:09:37.720 --> 0:09:39.760 nothing wrong. That's amazing. 0:09:39.840 --> 0:09:42.120 It's also kind of crazy to think that people would 0:09:42.160 --> 0:09:43.839 have been looking at that and just been like, oh, well, 0:09:43.840 --> 0:09:47.120 I guess demand for veterans benefits is lower than we 0:09:47.200 --> 0:09:49.360 thought it would be. But actually it was a tech issue. 0:09:49.400 --> 0:09:51.880 I think what they said was demand for them doing 0:09:51.880 --> 0:09:54.360 it online is low. These veterans must. 0:09:54.120 --> 0:09:56.240 Not have access, which is absolutely not true, or they 0:09:56.240 --> 0:09:58.520 can't figure out the computer when in fact, yeah, I 0:09:58.520 --> 0:10:00.640 mean them couldn't figure out the software. 0:10:00.679 --> 0:10:02.520 They would tell them. And they told this guy Dominic, 0:10:02.520 --> 0:10:04.480 the veteran that they interviewed, They kept telling him it's 0:10:04.559 --> 0:10:06.040 user error. There's something wrong with you. 0:10:06.840 --> 0:10:09.160 Can I ask you a question about that bidding process? 0:10:09.240 --> 0:10:14.760 When I hear government goes out to bidders for a website, 0:10:15.160 --> 0:10:21.280 I imagine these like big sort of faceless offices in Roslin, Virginia, 0:10:21.800 --> 0:10:24.199 where you know, people get like chop salid for lunch 0:10:24.240 --> 0:10:25.959 and stuff like that, and there's like three or four 0:10:26.000 --> 0:10:27.800 of them and they sort of like rotate who wins 0:10:27.840 --> 0:10:30.439 the bid and stuff like that. How competitive, like is 0:10:30.480 --> 0:10:33.800 it would a typical rfp auction or bidding process be 0:10:33.840 --> 0:10:37.640 and how hard would it be forcea a more agile, 0:10:37.840 --> 0:10:40.840 innovative firm or someone wanting to do it differently to 0:10:41.440 --> 0:10:43.600 break into the pursuit of that contract. 0:10:43.760 --> 0:10:46.920 That's changing a lot. So now you've got a set 0:10:46.960 --> 0:10:49.760 of vendors. I think there's about thirty five of them 0:10:50.040 --> 0:10:52.400 in this thing called the Digital Services Alliance, which are 0:10:52.559 --> 0:10:55.520 smaller companies that work in a more sort of agile, 0:10:55.600 --> 0:10:58.880 user centered way, and you know, they're just a lot 0:10:58.960 --> 0:11:02.280 smaller than the Beltway, so there's still a tiny portion 0:11:02.880 --> 0:11:06.319 of the market, but you know, they're really viable businesses, 0:11:06.400 --> 0:11:09.000 and people in government can contract out to them. In fact, 0:11:09.000 --> 0:11:10.959 I think there's a mechanism by which you can put 0:11:11.000 --> 0:11:13.800 an RFP out just to those companies. But what has 0:11:13.840 --> 0:11:17.560 happened in the past is like Dave can pile onto this. 0:11:17.679 --> 0:11:21.080 Like if you were bidding out, say you know benefits 0:11:21.120 --> 0:11:23.880 system in a state, and pick your benefit. Usually they 0:11:23.920 --> 0:11:28.200 would require in the RFP that the bidder had to 0:11:28.240 --> 0:11:32.240 have done a similar system in a different state in 0:11:32.320 --> 0:11:35.280 the same category, which means that right there it limited 0:11:35.280 --> 0:11:39.000 to like three vendors, and so you were never going 0:11:39.080 --> 0:11:42.880 to get any disruptors because they were boxed out in 0:11:42.920 --> 0:11:44.080 the very first step. 0:11:44.200 --> 0:11:47.040 Just to maybe jump in. That's exactly right, and if 0:11:47.080 --> 0:11:49.440 you I mean, to some extent, it's rational. If you 0:11:49.480 --> 0:11:52.239 are thinking about this as someone who's not a technologist, 0:11:52.360 --> 0:11:55.640 you run an agency, and you're thinking, we need to 0:11:55.679 --> 0:11:58.000 do this very large project. It's going to take a 0:11:58.120 --> 0:12:01.640 very long time. It we really need it to go right. 0:12:02.320 --> 0:12:06.760 It is hard to think about, well, why wouldn't I 0:12:06.840 --> 0:12:10.640 look for a vendor who's done this five, ten, fifteen times. 0:12:10.640 --> 0:12:13.840 The problem is, as Jen said, well, I think there's 0:12:13.840 --> 0:12:17.120 two problems in that. One is you've now very very 0:12:17.200 --> 0:12:20.400 drastically narrowed your vendor pool, so there's a lot less competition, 0:12:20.800 --> 0:12:25.200 and two you also may have too big of a project, 0:12:25.280 --> 0:12:27.680 Like really, what level of attraction are you trying to 0:12:27.720 --> 0:12:30.120 work towards? And is it the case that really you 0:12:30.200 --> 0:12:33.439 need someone or a vendor who's worked on this specific 0:12:33.640 --> 0:12:36.280 type of system in another entity that it works exactly 0:12:36.360 --> 0:12:38.959 the same way and all of those details of the 0:12:39.000 --> 0:12:42.880 exact same program, the exact same structure of agencies. Probably not, 0:12:43.040 --> 0:12:47.600 because probably it's about more generically software. But if you 0:12:47.679 --> 0:12:50.520 really focus only on the track record there, you kind 0:12:50.520 --> 0:12:54.680 of get stuck, as Jema was saying, in this trap 0:12:54.760 --> 0:12:57.839 where there's only a small number of vendors who can 0:12:57.880 --> 0:13:01.240 meet those criteria, even if you think they're And it's 0:13:01.280 --> 0:13:03.120 hard because the other thing I would say is as 0:13:03.200 --> 0:13:05.640 much as there are new vendors, and maybe Jenny can 0:13:05.640 --> 0:13:08.280 speak more of this, but it's also the case that 0:13:08.320 --> 0:13:10.280 if you have a new vendor that wants to work 0:13:10.320 --> 0:13:12.160 in a more agile way, wants to work in a 0:13:12.160 --> 0:13:16.360 more use this CRD way, but they are bidding on 0:13:16.440 --> 0:13:19.880 an RFP that is structured in a very waterfall top 0:13:20.000 --> 0:13:22.920 down meet other requirements, and that's what success looks like. 0:13:23.160 --> 0:13:26.280 It's also not going to be successful because you're not 0:13:26.320 --> 0:13:30.040 giving them any space to say, oh, well, I know 0:13:30.120 --> 0:13:33.840 your requirement says a way to ender phone numbers using 0:13:34.040 --> 0:13:37.000 the key the mouse, but when we tested it with 0:13:37.080 --> 0:13:40.840 people like people, turns out really don't like using the 0:13:40.920 --> 0:13:43.520 mouse to click numbers to enter digits. They prefer to 0:13:43.600 --> 0:13:46.880 use the keyboard. Could we change that requirement? And honestly, 0:13:47.120 --> 0:13:50.040 I mean, that's an extreme example, but a lot of 0:13:51.160 --> 0:13:52.760 and there's a lot of what Jen's book is about 0:13:52.880 --> 0:13:56.200 and what resonated with me in it is that in 0:13:56.679 --> 0:13:59.640 doing these things and building software, you learn those details 0:13:59.640 --> 0:14:02.040 that you never could have gotten right up front. And 0:14:02.120 --> 0:14:07.640 so if you make success defined by implementing everything as 0:14:07.679 --> 0:14:11.600 we understand it before we ever get started, you almost 0:14:11.720 --> 0:14:14.200 ensure that you're not going to get what you really 0:14:14.240 --> 0:14:17.559 wanted because you're not allowing any information after that cut 0:14:17.600 --> 0:14:19.800 off point, right when you might be getting to the 0:14:19.800 --> 0:14:22.160 point where in doing it you're gonna learn a lot. 0:14:39.280 --> 0:14:41.320 I definitely want to ask you more about that top 0:14:41.400 --> 0:14:44.800 down waterfall nature of how a lot of these software 0:14:44.840 --> 0:14:47.760 projects get commissioned. But before I do, I'm just curious 0:14:47.840 --> 0:14:51.600 about the actual vendors because I'm sort of imagining these 0:14:51.640 --> 0:14:57.200 like big faceless offices. But what types of vendors are 0:14:57.240 --> 0:14:59.720 we talking about? And I'm trying to think how to 0:15:00.120 --> 0:15:03.680 use those kind of diplomatically. But I can imagine a 0:15:03.800 --> 0:15:08.000 company that is putting itself out bidding for government contracts 0:15:08.120 --> 0:15:13.360 on software development, Like maybe they're not as experimental or 0:15:13.440 --> 0:15:17.320 cutting edge as a software company that's, you know, somewhere 0:15:17.360 --> 0:15:21.920 in San Francisco and it's building like new exciting products 0:15:21.640 --> 0:15:25.240 for the corporate world. Although we've also done an episode 0:15:25.240 --> 0:15:29.400 on how bad internal corporate tech can be, but it 0:15:29.520 --> 0:15:32.680 seems like it might be a little bit more I 0:15:32.720 --> 0:15:37.080 hesitate to say boring, but sort of basic playing it safe. 0:15:37.160 --> 0:15:39.520 Well, you know one thing that I will say, like 0:15:39.600 --> 0:15:43.960 I hear both incredible frustration and even anger from folks 0:15:44.000 --> 0:15:46.720 that have to hire these companies. They're just like, there's 0:15:46.720 --> 0:15:50.400 no alternatives. This has gone badly, it will go badly. 0:15:50.440 --> 0:15:52.880 They kind of know that, Like I had this woman 0:15:53.040 --> 0:15:57.120 in a major state. I won't say which who. When 0:15:57.120 --> 0:15:59.080 I was said, you know your project's going to fail, 0:15:59.160 --> 0:16:01.440 she said, do you think we don't know that? The 0:16:01.560 --> 0:16:04.720 last seven projects have failed. So there's this a huge frustration. 0:16:05.240 --> 0:16:11.560 But there's also this real feeling that the company in 0:16:11.600 --> 0:16:15.280 San Francisco with its forty people who make consumer software 0:16:15.760 --> 0:16:18.560 like great, people love using that software, but they don't 0:16:18.600 --> 0:16:21.880 understand us, right, They're not going to actually understand the 0:16:21.880 --> 0:16:25.040 constraints of government. So we have to go with these 0:16:25.040 --> 0:16:29.120 big companies. And they're not wrong in the sense that 0:16:29.240 --> 0:16:32.880 the constraints of working with government are real. There's a 0:16:32.960 --> 0:16:35.960 huge compliance burden that forty person company in Sum Francisco 0:16:36.640 --> 0:16:40.520 mostly does not want these jobs, Like there's so much 0:16:40.560 --> 0:16:44.520 that goes into getting the you know, getting the bid, 0:16:45.360 --> 0:16:48.160 you know these it's sort of famously said that these 0:16:48.200 --> 0:16:51.760 companies know how to get the get the work, not 0:16:52.320 --> 0:16:54.760 how to deliver on the work. That's not probably what 0:16:54.800 --> 0:16:58.400 a fordy person company wants to do, but you know, 0:16:59.000 --> 0:17:02.320 it's true that you have to deliver on stuff that 0:17:02.440 --> 0:17:06.840 is mind bogglingly complex. Like when we're working in unemployment 0:17:06.840 --> 0:17:12.200 insurance again during the pandemic, my colleague was talking with 0:17:12.400 --> 0:17:15.760 the claims processors like week over week, and we're trying 0:17:15.800 --> 0:17:18.520 to like dissect it and figure out what's going wrong 0:17:18.600 --> 0:17:21.600 and like clear this backlog. And one of these guys 0:17:21.680 --> 0:17:24.600 keep saying, well, I'm not quite sure about that answer. 0:17:24.680 --> 0:17:26.560 I'm the new guy. I'm the new guy. And she 0:17:26.640 --> 0:17:29.600 finally says, how long have you been here? And he says, 0:17:29.680 --> 0:17:33.119 I've been here seventeen years. The guys who really know 0:17:33.200 --> 0:17:35.879 how this works have been here twenty five years or more. 0:17:36.440 --> 0:17:40.880 So think about, like, you know, going from doing some simple, cool, 0:17:41.080 --> 0:17:44.720 you know tech app, you know, easy consumer app, to 0:17:45.200 --> 0:17:48.640 trying to build or fix or improve upon a system 0:17:49.240 --> 0:17:52.440 that is so complex that it takes twenty five years 0:17:52.520 --> 0:17:56.320 to learn how to process a claim. That's that's sort 0:17:56.320 --> 0:17:57.919 of I think what needs to be on the table 0:17:57.960 --> 0:18:00.680 as part of this agenda is not just like can 0:18:00.720 --> 0:18:03.000 the tech be better? But can we go back and 0:18:03.000 --> 0:18:07.440 simplify the accumulated like ninety years of policy and process 0:18:07.840 --> 0:18:09.560 that's making that so hard to make? 0:18:09.920 --> 0:18:10.120 Well? 0:18:10.240 --> 0:18:11.880 Why do we sort of back up? 0:18:11.920 --> 0:18:12.080 Then? 0:18:12.119 --> 0:18:14.800 And I again kind of steal Tracy's question because I 0:18:14.840 --> 0:18:17.399 had the same thought, what is it about government buying 0:18:18.040 --> 0:18:23.800 that creates these massive waterfold RFPs in which you can't 0:18:23.880 --> 0:18:27.719 you know, in which a like create these constraints that 0:18:27.760 --> 0:18:31.200 the software developer would not necessarily have with a private buyer, 0:18:31.640 --> 0:18:35.879 and b puts government agencies in a position where they 0:18:35.920 --> 0:18:37.359 will say to you, of course, we know it's going 0:18:37.440 --> 0:18:40.840 to fail, which again presumably that actually probably does happen 0:18:40.880 --> 0:18:43.320 in the private sector too. Or but what is what 0:18:43.359 --> 0:18:45.800 are the to some extent, but what are the dynamics 0:18:45.920 --> 0:18:48.560 that make these things like so hard to change and 0:18:48.640 --> 0:18:52.040 so hard to you know, rip, you know, come up 0:18:52.080 --> 0:18:52.760 with a new system. 0:18:54.280 --> 0:18:56.320 Well, this is a good Dave question, but I'm gonna start. 0:18:57.600 --> 0:18:59.960 You know, I really spend a lot of time reflect 0:19:00.200 --> 0:19:02.200 on this, you know, I had. I've had sort of 0:19:02.240 --> 0:19:04.080 three years since I stepped down from Code for America. 0:19:04.119 --> 0:19:05.720 I just like sat in a room and well, I 0:19:05.760 --> 0:19:08.760 interviewed people and thought about it, and I thought about 0:19:08.760 --> 0:19:12.080 my own experiences, and I think that there's a deep 0:19:12.119 --> 0:19:16.040 seated culture in government where the policy people are the 0:19:16.080 --> 0:19:20.840 important people. They do the important stuff, and technology digital 0:19:21.000 --> 0:19:23.760 is just part of implementation, which is at the not 0:19:23.800 --> 0:19:27.520 just the bottom of like a software development waterfall, it's 0:19:27.520 --> 0:19:31.760 the bottom of a big, rigid hierarchy and which information 0:19:31.920 --> 0:19:34.280 and power and insights only flows from the top to 0:19:34.320 --> 0:19:38.119 the bottom. And so it's problematic in part because the 0:19:38.160 --> 0:19:41.800 people who are doing the tech are really just sort 0:19:41.840 --> 0:19:46.560 of downstream of everything else and the power and ability 0:19:46.640 --> 0:19:49.480 and willingness to step up and say, hey, like we 0:19:49.680 --> 0:19:53.560 probably shouldn't do those sixy seven hundred requirements, we should 0:19:53.560 --> 0:19:57.440 probably focus on these two hundred. Get that out the door, 0:19:57.480 --> 0:19:59.960 and then you know, adage cases as later like that's 0:20:00.000 --> 0:20:03.360 it's just not there's no permission really to say that, 0:20:03.480 --> 0:20:07.439 and until we you compare that with say, we'll call 0:20:07.440 --> 0:20:09.879 it metaphysical silicon valleys. That I don't mean actually like the 0:20:09.920 --> 0:20:13.680 silicon valley, Like the people I write code started the companies, 0:20:13.680 --> 0:20:16.879 they're in power, they're at the top, Like compliance like 0:20:17.400 --> 0:20:23.160 is below them. In sort of the DC government hierarchy, 0:20:23.240 --> 0:20:26.640 compliance is like way above, right, Like policy is way 0:20:26.680 --> 0:20:29.440 above and there's not this sort of build measure learn 0:20:29.560 --> 0:20:33.960 cycle the Dave was referring to, where you know, people 0:20:34.000 --> 0:20:36.720 are learning from each other as they're doing the work 0:20:36.760 --> 0:20:40.040 and that technology. You know, certainly the policy has an 0:20:40.040 --> 0:20:42.440 impact on the technology, but the building of the technology 0:20:42.480 --> 0:20:45.400 needs to have an impact on the policy, like that 0:20:45.680 --> 0:20:50.680 fundamental culture is something that needs to be like looked 0:20:50.680 --> 0:20:54.120 at and called out, and where it's not happening, where 0:20:54.160 --> 0:21:01.639 there's actual conversation, you know, dialogue beats right. Where that's happening, 0:21:01.680 --> 0:21:04.280 we should be lifting that up and saying this is 0:21:04.320 --> 0:21:07.520 possible within government because that's like the foundation I think 0:21:07.520 --> 0:21:12.119 of getting away from these mega projects that fail. But 0:21:12.200 --> 0:21:14.040 Dave you you are you're going to disagree with me 0:21:14.040 --> 0:21:14.919 on this, and I'm gonna love it. 0:21:15.000 --> 0:21:17.119 I think that's I actually do think that's right. I 0:21:17.160 --> 0:21:21.480