Here's the low down about low-code tools
In this episode, we talk about the termination of Dr. Margaret Mitchell, founder and lead of the Google Ethical AI team, Australia’s law requiring Google and Facebook to pay news publishers, and concerns over Clubhouse security. Then we are joined by Rahat Chowdhury, co-founder of Whimser, to chat about the large funding rounds low-code platforms have been receiving, what this means for developers, and where these tools might fit in our own development.
Saron Yitbarek is the founder of Disco, host of the CodeNewbie podcast, and co-host of the base.cs podcast.
Josh Puetz is Principal Software Engineer at Forem.
Rahat is a second career developer who comes from a background in customer support management. He is a mental health advocate, entrepreneur, and teacher who enjoys spending his free time mentoring new developers.
[00:00:13] SY: Welcome to DevNews, the news show for developers by developers, where we cover the latest in the world of tech. I’m Saron Yitbarek, Founder of Disco.
[00:00:21] JP: And I’m Josh Puetz, Principal Engineer at Forem.
[00:00:24] SY: This week, we’re talking about the termination of Dr. Margaret Mitchell, Founder and Lead of the Google Ethical AI Team, Australia’s law requiring Google and Facebook to pay news publishers, and concerns over Clubhouse security.
[00:00:38] JP: Then we’re joined by Rahat Chowdhury, Co-Founder of Whimser, to chat about the large funding rounds low-code platforms have been receiving, what this means for developers, and where these tools might fit in with our own development.
[00:00:49] RC: The majority of my team are all front-end devs. So we don’t have anyone kind of focused on the back end and we haven’t had to reach out to do that because of the service.
[00:00:58] SY: We’re starting off this episode with an update about the ongoing saga that is the fallout from Google’s firing of one of their lead AI ethics researchers, Dr. Timnit Gebru, who criticize the company’s efforts to hire underrepresented people in tech, as well as biases in their AI. Please listen to Season 2, Episode 7, where we go more in depth about the events that led up to the termination of Dr. Gebru with Julien Cornebise, an honorary associate professor at University College London and a former researcher with DeepMind, Google’s AI lab. In Episode 5 of this season, we talked about how Dr. Margaret Mitchell, the founder and lead of the Google Ethical AI Team, who callout the team with Dr. Gebru was logged out of her corporate account after she’d written an email to Google Press criticizing the termination of Dr. Gebru. We’ll put Dr. Mitchell’s letter in our show notes. A Google spokesperson said that Dr. Mitchell’s corporate access had been locked because she had downloaded and shared sensitive documents with external accounts and that they were currently investigating the incident. Well, after five weeks of being locked out of her account, Google officially fired Dr. Mitchell on Friday, February 19th. Also on that Friday, Jeff Dean, the head of AI at Google, wrote in a memo obtained by Axios that the internal investigation of the termination of Dr. Gebru was complete saying, “I heard and acknowledged where Dr. Gebru’s exit signified to female technologists, to those in the black community and other underrepresented groups who are pursuing careers in tech and to many who cared deeply about Google’s responsible use of AI. It led some to question their place here, which I regret.” The memo also states that the company has also updated policies in terms of diversity and inclusion, as well as the publishing of research. However, the details of the investigation have yet to be released. The termination comes right after Google announced a reorganizing of its teams working in AI ethics and that VP of engineering, Marian Croak, would be leading this part of the organization. Dr. Gebru tweeted about the announcement, “Announcing a new org by a black woman as if we’re all interchangeable while harassing, terrorizing, and gaslighting my team and doing absolutely zero to acknowledge and redress the harm that’s been done is beyond gaslighting.” It should be noted that we’ve reached out to Dr. Gebru and Dr. Mitchell to come on the show. All right.
[00:03:26] JP: Thank you for that recap. Holy crap!
[00:03:28] SY: Sorry. My pop filter fell down.
[00:03:31] JP: That’s such a hot recap. It just blew the pop filter right off.
[00:03:36] SY: So Josh, first of all, let me start by asking you this. Were you surprised?
[00:03:41] JP: I think I was surprised that it took so long. They locked Dr. Mitchell out of her email for five weeks, and then they fired her. Did they fire her out of week one so she just couldn’t check her email? Of course not. But it’s ridiculous that that would take that long.
[00:03:55] SY: Yeah. Yeah.
[00:03:56] JP: It smacks. They were just looking for a reason to fire her or trying to come up with some reason. I’m not surprised she was fired. I’m very surprised that the whole situation is taking place. And yet I’m not. These ethicists are doing great work and they’re pointing out ethical problems with the products and the actions that Google is taking. And that’s a tricky balance and Google seems to have been completely unprepared for employing people that are going to call out the ethical problems with its work in public.
[00:04:30] SY: Yeah. I think that to me was the surprising part because they hired these two women, one, a black woman, and obviously big tech has been criticized for diversity and inclusion for many, many years. So this isn’t anything new. What did you think was going to happen? Of course, you’re going to be given some feedback and of course you’re going to be held to a certain standard being Google and having the influence and the power that you have. So I was very surprised that this was the reaction. We don’t know the investigation and we probably will never know exactly what happened, in terms of at least what Google thinks happened and what the reasons are. We won’t have those details, but it’s just confusing, I guess. And it’s also kind of confusing to have a second firing after the backlash of the first firing. That was the thing that was like, “Really? You really think that’s a good idea?” And you’d think after all of the support Dr. Gebru has gotten and after all of the conversations and the rage and the backlash that they would kind of act right and that they wouldn’t kind of do it again. And maybe they waited five weeks hoping it would die down. Maybe that was the strategy, but it was very surprising that they did exactly the same thing again to the co-lead of all people, like to Dr. Gebru’s equivalent, they did that again. So yeah, I was very surprised.
[00:05:54] JP: I think the two big themes I’ve taken away from this is that Google has a diversity hiring and retention and work environment problem, like many large tech companies. However, Google doesn’t want to talk about it. They don’t want to engage with it. They just want to shut down all discussion of it. Period. The second problem they have is the idea of hiring ethicists within their company and letting those ethicists do work that could potentially criticize the industry at large and Google in particular. Google hires lots of other researchers, but they’re doing computer science research, machine learning research, and those kinds of research papers, you’re not going to publish a machine learning research paper that somehow criticizes Google. You’re going to publish a machine learning paper that posits a cool new technology or an interesting way of doing things. But nobody explained to Google that the practice and field of ethicists is to call actions into questions.
[00:07:03] SY: That’s literally the job.
[00:07:03] JP: It’s literally the job to call out whether something is ethical or not. I don’t know if Google thought like, “Cool, we’ll get some ethicists and they’ll publish papers saying Google’s awesome and what they’re doing is ethical and everything is great.” It’s like nobody explained what ethics is to Google before they decided to start up this team.
[00:07:21] SY: I don’t know what they thought was going to happen. I don’t know what the future of ethics is going to be like. I can’t imagine anyone really taking it seriously after this.
[00:07:29] JP: Certainly not at Google.
[00:07:30] SY: Even if they had not exactly, like even if they had some superstar ethicist being brought on board, I don’t know if it would really mean anything. I think this has really damaged their reputation in the AI community and just makes everything so suspicious and just very untrustworthy. So I just don’t understand how they’re going to repair their reputation after this.
[00:07:51] JP: Or if they even bother.
[00:07:55] SY: Maybe they just don’t care.
[00:07:56] JP: I mean, really, would you believe an ethics report from Google now about their AI or any AI in general? No.
[00:08:01] SY: Absolutely not. No. Well, we’ll see what happens and we’ll see what Marian Croak, the new woman who’s going to be leading AI ethics. We’ll see what she does and we’ll see how this all plays out.
[00:08:13] JP: I’m also very interested to see where Dr. Gebru and Dr. Mitchell land and what their next acts are.
[00:08:20] SY: I’m sure they’re in high demand.
[00:08:21] JP: Right. I’ll be paying much more attention to their research and their publications in the future than frankly I will to Google’s ethics AI team.
[00:08:29] SY: Right. For sure.
[00:08:32] JP: So this bit of news has been brewing for a while. Back in August of 2020, the Australian Competition and Consumer Commission was drafting a bill to be sent to parliament that will require both Google and Facebook to pay news publishers for news content that appears on those companies’ platforms in Australia. Now that law is expected to pass the Australian government in a couple of weeks. And after fighting hard to prevent this law’s passage, both Google and Facebook are striking deals with the country’s government. Google announced a three-year partnership with News Corp, the mass media corporation owned by Rupert Murdoch, which essentially owns all of the news publishers in Australia. Google is going to create a news showcase portion of its search engine that will feature publications, including the Wall Street Journal, The Times, and The Australian. The company will then share a portion of the ad revenue they generate from traffic going to these sites with News Corp. Facebook initially decided to take a different approach and they blocked any news publishers from appearing on their social media platform in Australia. And not just news publishers were blocked, but pages for Australian nonprofits, state health departments, and emergency services also vanished overnight. Afterwards, the company blamed the law’s definition of news as too broad and said it would restore public vital service pages, but not the links or articles from news publishers. However, after four days of Facebook banishment, the social media company struck a deal with the Australian government on Monday, February 22nd, and said it would restore news links and articles. Some of the concessions the Australian government made with Facebook after the company came back to the bargaining table is more time for the company to negotiate deals with individual news publishers. Also, if digital platforms contribute significantly to Australia’s news industry, they could potentially opt out of the law. And one other nugget that was a sticking point in negotiation was whether Facebook had to agree to binding arbitration in these negotiations. And that requirement has been somewhat relaxed. So I found this story fascinating because this is a huge media landscape story. It definitely affects developers on the web, which we’ll get into in a second. But this is my old US centralism taking hold. I don’t pay a lot of attention to the news media world outside of the United States. And that’s my fault. And this whole situation happening in Australia, I hadn’t been paying attention to it, and it’s a big deal. We’re starting to see other governments do this.
[00:10:57] SY: It’s a really big deal.
[00:10:59] JP: Now we’re starting to see other governments take a look at Australia. The Canadian government, the Irish government have both been kind of looking at this and thinking, “Huh? I wonder if we could get Facebook and Google to pay our news publications for their stories.”
[00:11:10] SY: It’s fascinating. I can’t believe Google and Facebook caved, to be honest. When I was first reading about this story, I was thinking, “Well, Google and Facebook kind of have all the cards. So what are you going to do?” You know what I mean?
[00:11:21] JP: Yeah.
[00:11:21] SY: Because, ultimately, that is a great way to generate traffic and there is a huge benefit of being on Google Search. I mean, just think about the industry of SEO and all of our efforts to get on that first page of Google. There’s a huge amount and value there. So I just assumed that Google and Facebook had all the leverage in that conversation and that they wouldn’t cave, and I’m very surprised they decided to. And I’m surprised that they decided to now, Google and Facebook have been around for a very long time. And this has been a conversation for a very long time, right? If you’re the one that’s going to be featured on their page, should you get some of the ad revenue? How do you get a chance to benefit from it besides the traffic? That’s been a conversation that’s been on the table for a while. And I’m surprised that it happened now that these concerns, these calls were taken seriously now. And I’m kind of curious about why that is. If Australia had kind of had this conversation five years ago, would they have won? Would they have been able to strike a deal? I’m curious about the environment and why this is happening at this point.
[00:12:28] JP: So one analysis I was reading had said that Google gets a lot more of its traffic from news than Facebook does. And I think that kind of might explain the two companies’ differing approach. I’ve also read some analysis that pointed out that Google is really used to playing by different rules in different countries. You take a look at Google’s approach in China, Google’s approach in authoritarian countries and they’re very adept at molding what services they offer and what rules they play by in order to remain in those various countries. So that might’ve played into it a little bit. Facebook on the other hand, Facebook doesn’t give to reps and I don’t think a lot of Facebook’s traffic comes from news per se. So that might’ve been a little bit of it. I also read an analysis that said after I think it was four days of Facebook taking down all these news links that news publishers in Australia were seeing drastically reduced traffic numbers, as exactly as you would expect. And that might’ve gotten the Australian government at the behest of these news publishers back to the negotiating table. I agree with you. It seems wild to me that either one of these companies would feel like they have to play by the rules that the Australian government is setting out. The other big thing that was fascinating to me about the story was that it’s not so much like the law has been passed and now Facebook has to pay X, Y, Z. The amounts they have to pay, whether they have to pay, the law says that all has to be spelled out in arbitration, which is not strictly a legal trial and is typically used to give, when you have two parties and one party has way more power than the other party, arbitration is a way to say, “Okay, we’re just going to let a third party decide and we both have to agree to it.” Obviously, Facebook and Google are the more powerful parties in this case and they’re not so wild to agree to that. So I’m really surprised by this whole thing.
[00:14:21] SY: What do you think this means for the relationship between Google, Facebook and the media in other countries?
[00:14:29] JP: I’ve really worried about it, to be honest. I think this is a complicated question. At the root of it, if you’re a journalistic organization and you’re producing news, do you have a right to all the traffic that news story will ever get no matter where it is? And I think that’s the crux of the question. And I am worried that the set of precedent that the news industry, which frankly is hurting for exposure and income, will lean on this as a crutch to maintain themselves.
[00:15:02] SY: One hundred percent.
[00:15:03] JP: The bigger thing this makes me worry about is, and this is something Facebook had argued in their initial briefs, this kind of fundamentally changes the rules of the internet. Many, many years ago, when the web was first getting started, you would see large companies complaining saying, “Hey, blogger, you need to take this link down to this article or this page that we posted. We won’t allow you to link to our content.” Web developers laugh at that. “You can’t control what I put on my own web page and you can’t control what I link to. That’s the nature of the web. We can link to whatever we want. If you don’t like it, put a password in front of it or lock it down somehow, but you can’t sue me. You can’t prevent me from linking to pages on the internet.” But that’s exactly what these laws would do. They would say, “Facebook, you can’t link to The Wall Street Journal without giving them money.” That really flips the relationship and changes one of the fundamental rules of the internet.
[00:16:05] SY: That’s a really interesting point. Yeah. And that historical context, it does make it really concerning. You know what? I was thinking about the ramifications, I was thinking about it more from a business perspective and I’m not trying to say that we need to be concerned about Google’s profit margins, but it is kind of interesting to think about how these conversations might happen in other countries. Obviously, the bigger publications have more leverage, but what happens to the smaller publications? Do they kind of somehow make it to the negotiation table? And if Google and Facebook are striking deals with all these publications, one, how does that change our experience of the internet, if it does? And then two, what does happen to the business side of Google and Facebook? And kind of related to that, one thing that Google has been doing a lot over the years is not just linking to these websites, these journalism publications, but they’ve also been just straight taking the data and posting it on the homepage. So if The New York Times is getting a cut of the revenue, then what happens to hotels.com when the hotels’ listings are scraped and then put on the actual search page itself? You know what I mean? This can really go on. This can unravel for a while, and I’m really interested to see how this affects Google’s whole business model in the future.
[00:17:35] JP: What other things we should point out that I found interesting about the law is that it’s all concerned with money going to the news organization companies and not the news organizations or their journalists in particular. So frankly, Rupert Murdoch is getting all this money. News Corp is getting all this money. It’s not going to the newsroom of The Wall Street Journal or The Australian or The Times. And you bring up a good point about smaller news organizations. News Corp, they’re falling under hard times. They’re trying to maintain a profitable business. But they’re not the little guy in this equation by any means. And I think that really harms them.
[00:18:16] SY: Right. That is such a valid point. The idea of what they get paid for their articles definitely sounds great. Just on the surface, it sounds like, “Oh, we’re really standing up for journalism and we’re supporting these writers.” But if you think about literally where the money is going and whose pockets they end up in, it’s a different story.
[00:18:34] JP: Right. I would not be surprised if we see someone in the United States, Senate or House, propose a law like this or start to talk about that.
[00:18:44] SY: It’s coming.
[00:18:44] JP: Yeah. I think it’s absolutely coming.
[00:18:46] SY: It’s definitely coming.
[00:19:05] SY: RudderStack is the Smart Customer Data Pipeline. Easily build pipelines connecting your whole customer data stack, then make them smarter by ingesting and activating enriched data from your warehouse, enabling identity stitching and advanced use cases like lead scoring and in-app personalization. Start building a smarter customer data pipeline today. Sign up free at rudderstack.com.
[00:19:28] JP: Are you looking to build a chat for your next big project? Save time building in-chat, voice, and video for mobile and web applications with Sendbird. Get to market faster with Sendbird’s UIKit, pre-built UI components, best-in-class documentation, and support for developers. Join companies like Reddit, Delivery Hero, Yahoo Sports, and Hinge. Start your free trial today at sendbird.com/devnews.
[00:19:53] SY: So on Monday, a developer was banned from Clubhouse after creating a third-party app that allowed them to share their login with other users and to stream audio from the invite-only apps chat rooms, causing some to say that they had “hacked Clubhouse”. However, as cybersecurity expert and CEO of Internet 2.0, Robert Potter, astutely pointed out in a tweet, “At the technical level, the user who shared his account through the open source platform, they didn’t really hack anything aside from himself. It’s likely a violation of the terms of service, but not what you would call a hack. The real problem was that folks thought these conversations were ever private.” So Clubhouse said that they’ve put in new safeguards to ensure that something like this doesn’t happen again. I don’t know exactly how they can ensure something like this isn’t happening because, I mean, first of all, the idea that anyone thinks that their streamed conversations are private is just ridiculous.
[00:20:54] JP: Wait. My internet party line isn’t private. What?
[00:20:57] SY: Yeah.
[00:20:58] JP: I thought it was just me and Elon Musk having a private conversation. Who would have thought that?
[00:21:02] SY: This is such a ridiculous assumption. I don’t know who thought that was a good idea, like who believed in that.
[00:21:09] JP: Maybe it’s because of the exclusivity or perceived exclusivity of Clubhouse. You have to be invited to be on Clubhouse. So maybe that led to it.
[00:21:15] SY: Maybe. But I mean, ultimately, it’s audio. I can just pull up my recorder, like my phone up to another phone and record the conversation and put it on YouTube or SoundCloud. If you say something out loud via the internet, it can be recorded, and I don’t think that should be a revelation.
[00:21:38] JP: Right. It’s the same thing we see with pirated movies. Somebody can smuggle a movie camera into a theater and point it at the screen. As long as that final data stream has to go to your eyeballs or your ears, someone can record it.
[00:21:55] SY: And at least in the movie theater, there’s like the ushers, it’s pretty obvious if you bring out a camcorder. The audio is probably really low quality. It’s very different with Clubhouse, right? No one sees what you’re doing. The audio quality in Clubhouse is already really bad.
[00:22:09] JP: Right. It’s kind of janky to begin with.
[00:22:13] SY: It’s already bootlegged, right? It’s already bootlegged. So yeah. I found this to be absolutely hilarious. And yeah, I just don’t know exactly what Clubhouse can do to prevent something like this from happening. I assume that having it, the iPhone only makes it probably easier. They’re not on the web, right? I think they’re just iPhone.
[00:22:31] JP: They’re just iPhone, but they have an API. That’s what holds all these hacks. People are writing their own iOS clients to Clubhouse and then putting them up on GitHub and you can just put your own credentials in there and then share the audio with other people. You get banned pretty quickly.
[00:22:46] SY: Yeah. I don’t know why they would have an API at all at this point. That just sounds like a bad idea.
[00:22:51] JP: This totally reminded me of when people were screenshotting Snapchat pictures and there was, “Oh my God! It’s not private?” No, it’s not private. Anything you put on the internet is not private.
[00:23:01] SY: Yeah. Yeah, exactly. I do love the ingenuity of this developer though. Just the idea that it was supposed to be this invitation. I mean, it is this invitation-only system. And through a little bit of creativity, they got around it and made it available to all people. I think it was just a really interesting way to get around the exclusivity of the system.
[00:23:23] JP: There’s been a lot of interest in doing this in China, especially, because as I understand it, the Chinese government has banned Clubhouse because they, everybody else seems to be able to record it, but the Chinese government can’t monitor it. And so it violates their censorship laws. But on the other hand, if you ever do get into Clubhouse and take a look at the rooms that are popular, there are tons of Asian language channels that it’s really taking off in South Asia. So there’s a real interest among the Chinese to try to somehow get into Clubhouse or at least listen to these conversations. I think that’s kind of an interesting wrinkle that’s motivating a lot of these “hacks”.
[00:24:04] SY: So we know that it has to do with an API and we know that it has to do with sharing the developer’s login, but how did it actually work?
[00:24:13] JP: So I took a look at the code that was shared on GitHub. And what’s really interesting is you can actually do a search on GitHub for Clubhouse and find other projects that are doing this. And the general approach seems to be you write an iOS app that ties into the Clubhouse API that authenticates you just like the official Clubhouse client does and then you post this project with your own identifiers. So you’re basically distributing a Clubhouse client that has you as the developer already logged into it and then other people can go ahead and either listen to it or pipe it to iOS audio recording stuff that the official Clubhouse client doesn’t allow. What I don’t understand is why the Clubhouse API isn’t more locked down. If they really don’t want people doing anything else with it, well, I have this API at all, you’d think would be encrypted to ridiculous levels, but it’s not.
[00:25:16] SY: I was going to say if it’s an invite-only platform, then clearly they’re trying to maintain some level of exclusivity. I mean, to be fair, there’s like millions of people on it. So how exclusive is it really? But still, if they’re trying to kind of create this feeling of being part of a club, I would not think they’d have an API at all, to be honest. I would think that they would want to really own and completely have that environment and that functionality entirely to themselves. So I’m surprised they even have an API.
[00:25:44] JP: Yeah. It seems like the API isn’t really public. The best way I could really equate it to is like let’s say I wrote a Gmail client and I put it on the web for everyone to use, but I baked my Gmail credentials into it. Now that wouldn’t be super exciting for me or for you, because all you’d be able to see are the emails that addressed me. But what if I did that with Clubhouse, you’d be able to look at basically anything. In Clubhouse, once you have an invite to Clubhouse and you’re in there, you can kind of go to almost any room. There are private rooms, obviously, but all the public rooms you can go visit.
[00:26:19] SY: Right. Right. Right.
[00:26:20] JP: I should also note that what I’ve seen from these different GitHub projects is that this is all mostly done by reverse engineering the app, listening on the wire to traffic that’s going back and forth between the Clubhouse application and their servers and none of it appears really encrypted.
[00:26:39] SY: If you want to hear more of our thoughts on Clubhouse, make sure to check out Season 3, Episode 5. There have been a slew of large funding rounds for different low-code platforms. The latest of which is Creatio, which just received $68 million in funding. Another local platform that has received a good amount of funding is OutSystems at $150 million, not to mention companies like Salesforce, Airtable, and Webflow, which have all become very popular household names. This investment into low-code platforms had us thinking about what this means for developers and where these platforms potentially exist in our own potential tool belts. So coming up next, we chat with Rahat Chowdhury, Co-Founder of Whimser, who uses low-code tools in his own development after this.
[00:27:46] Are you looking to build a chat for your next big project? Save time building in-app chat, voice, and video for mobile and web applications with Sendbird. Get to market faster with Sendbird’s UIKit, pre-built UI components, best-in-class documentation, and support for developers. Join companies like Reddit, Delivery Hero, Yahoo Sports, and Hinge. Start your free trial today at sendbird.com/devnews.
[00:28:12] RudderStack’s Smart Customer Data Pipeline is warehouse-first. It builds your customer data warehouse and your identity graph on your data warehouse with support for Snowflake, Google BigQuery, Amazon Redshift and more. Their SDKs and plug-ins make events streaming easy and their integrations with cloud applications like Salesforce and Zendesk help you go beyond event streaming. With RudderStack, you can use all of your customer data to answer more difficult questions and then send those insights to your whole customer data stack. Sign up free at rudderstack.com.
[00:28:48] SY: Here with us is Rahat Chowdhury, Co-Founder of Whimser. Thank you so much for being here.
[00:28:52] RC: Yeah. It’s awesome to be here.
[00:28:53] JP: So tell us about the company that you’ve co-founded, Whimser.
[00:28:56] RC: Whimser is a mental health based company. So it’s a mental health startup. We’re trying to use a cognitive behavioral therapy-based approach to self-help and eventually kind of get people to get used to the whole, I guess, therapy type deal, and eventually set them up with some actual professionals who can help them further than we can. We’re just trying to start the beginning of the journey. Just get them kind of started towards like, I guess, taking better care of their mental health.
[00:29:25] SY: Tell us about the stack you’re using to build this.
[00:29:28] RC: On the front end, we’re using React Next.js with Capacitor and Ionic to build out the mobile app on the website. We’re using Next.js as well. And on the back end, AWS Amplify. We’re kind of just using that to sort of take care of all of our database needs, authentication, things like that.
[00:29:46] JP: Very cool. So let’s talk a little bit more about that. Where do low-code tools fit into your stack? Why have you chosen them? And basically just talk to us a little bit about where low code fits into how you’re approaching development.
[00:30:00] RC: So I first started this kind of on my own before we kind of expanded the team. And what I wanted to do was make sure we’re kind of building the back end as easily as possible with the least amount of tech debt as possible. And knowing that I’m kind of like more of a front-end engineer, I was looking for something, I guess, more managed or easier to use. So I came across AWS Amplify. It takes a lot of the boxes I needed to make sure that whatever I was using was HIPAA compliant as well as easy to use and scalable. So the really cool thing that we can kind of do with Amplify is like they have this new admin UI and kind of just go in, click a few buttons, you’ve got a database, you’ve got like content that you can model right away without writing any code authentication that can be set up for you. The only time we’re kind of like actually coding is when we’re actually consuming all these services on the front end. But the back end is pretty much a bunch of clicks and we’re pretty much all set to go from there.
[00:30:57] SY: So paint us a picture of what it looks like to actually use these tools to integrate them into the rest of your development. What does that look like?
[00:31:04] RC: Essentially, I would spit up, I guess, like a back-end service or app in the admin UI, go in and I would kind of decide what do I want to do. Do I want a database? Do I want authentication? And starting, I guess, with the authentication, you automatically get access to stuff like Amazon Cognito. So you go in, you click authentication, and you can set up different attributes that you want for your user, like I want the user to be able to sign up with their email, with their password, to have like SMS verification. All that kind of stuff is kind of like baked in, just click a few buttons. You just choose kind of like the experience that you want the user to have. And it’s pretty much all done for you right there. You got them go in, you log in. There’s email verification that you get like a little confirmation code and stuff, like all of that kind of just works out of the box without much configuration.
[00:31:59] JP: Is this the first time that you’ve used low-code tools to this degree in a project?
[00:32:04] RC: Yeah, definitely. I was probably one of those developers who was like, “Oh no, I can’t do low code. I have to kind of do everything by myself.” So it took a little time for me to kind of get used to that. But what really kind of pushed me over to really using low-code tools is I wanted to make sure I got a product out. I wanted to make sure that I wasn’t just doing the same type of thing, I was doing over and over writing a whole authentication thing from scratch, again, like the fifth time or sixth time, or trying to do some basic functionality that I would write in every single app and it wouldn’t really be much different than what I’ve written before. So kind of all that stuff, being able to just go in and click a few buttons to have that was great because now I can kind of just like focus on the actual business logic of the app and actually get features out. We’re working like on the MVP we’re hoping to release soon. So just having a small team working on pretty significant and large app and not having to, I guess, worry about the logistics of like a whole back end that we have to create has been super helpful.
[00:33:09] SY: How do you think these tools can be best utilized by developers? Where is their place within developers’ tools and workflows?
[00:33:17] RC: Yeah. I think a developer might have their strengths, like I mentioned for my strength like write in front end. So I wasn’t a huge back-end person. So I was looking up a low-code tool for back end. And I think like folks who are probably more back-end focused can look for more front-end low-code tools to help them out too. When we’re really trying to maybe get out like a prototype or just a product playing to our own strengths and then using resources that help us fill in the gaps is probably one of the best ways to go about it. That’s kind of my mindset when I was going into using these.
[00:33:54] JP: What are some of the challenges that you’ve run into using a low-code tool?
[00:33:58] RC: One of the bigger challenges I think was really while it can be magical to just kind of have a lot of those things there, some understanding of, I guess, the underlying code is definitely to have. So that’s something that we kind of went in and tried to work on together. So I guess that’s both like a challenge and like sort of a good thing because it was a learning experience for us to learn a few more, new things about AWS and the overall back end services we were using.
[00:34:26] JP: I think one of the fears some developers have about low-code and no-code tools is that you’ll end up with an application that looks like everybody else’s application, because you are kind of using the same tools. So I wanted to ask, what has your experience been with the customization and extensibility of these tools? Are you finding that you are able to customize the output of these low-code tools? Are you finding that there’s room to do that? And basically talk me through that story. Especially as you being a front end developer, you probably have a lot of experience with, I guess, we would say high-code tools doing that kind of customization.
[00:35:05] RC: Yeah. I feel like with the customization aspects, at the basic level, since we’re kind of like getting some back end stuff, a lot of this stuff, authentication, database modeling, doesn’t change very much from app to app. So that’s been good to kind of have sort of onboard. One of our use cases is kind of like me using some filtering of like different what are called “cognitive distortions” based on like inputs from the user. So we know that AWS also kind of had this sort of a machine learning service that we could use to kind of like bucket all those different phrases under different categories to help the user out when they’re looking for different cognitive distortions to label their entries with. What we did there was we kind of got an introduction to it through the admin UI tool. And then from there, it led us to kind of like the documentation that was further in there to further customize it for our needs. So that’s what I kind of like about, I guess, the admin UI for AWS. They start you off with the basic set this up, get running, but then it's broken down further like if you need to customize it further, here’s some documentation to do that. And I think that’s something that a lot of local tools should probably make sure they have in there. The starting point would be great to have that low-code tool getting you there, but having that option for customization, like links and documentation, making that easy.
[00:36:34] JP: Have you run into any compatibility issues? I’m thinking specifically about using low code for the back end, like integrating into different authentication systems, different databases. Have you run into any problems where you’ve butted up against the edges of the tool in terms of compatibility with either services or even with the customizations you’re making?
[00:36:56] RC: So far, we’ve been pretty good at that. It’s pretty much served our needs. I think maybe there could be some issues down the road because we’re eventually planning to have different types of accounts. So right now we’re kind of just focusing on just regular users who want to use this app as a self-help type tool. Eventually we’ll want to have accounts specifically for like professionals as well who can come in. So I think probably just like trying to figure out the ins and outs of the cognito authentication service of how to differentiate between users is probably going to be one thing. I know they have some built-in stuff for that, but I think a lot of it is really just kind of like a learning curve that we just kind of have to overcome of like going in and figuring out what the, I guess, how like the API is structured for us to actually have different type of user accounts, things like that.
[00:37:50] SY: So there has been a lot of funding going towards low-code platforms. Creatio received $68 million. OutSystems has got $150 million. What’s your reaction to that? Do you feel like there is this big of an opportunity in low code?
[00:38:06] RC: Yeah, definitely, especially for someone like myself and my team. We haven’t really gotten to the point where we’re funded yet and we’re trying to bootstrap together our company and just having those tools available is great because I feel like different entrepreneurs who are just trying to start up a company or whatever can access these tools and create things faster, get product shipped out faster without having to, I guess, hire additional developers. Like you can just hire a front-end dev and use something like the admin UI. The majority of my team are all front-end devs. So we don’t have anyone kind of focused on the back end and we haven’t had to reach out to do that because of the service. It’s saved us time, saved us money and saved us just the aggravation I guess of like finding someone who is great on the back end and can help us with this.
[00:39:01] JP: So I’ll ask a question that I think some developers are probably wondering. Are you worried about the future of developer jobs if low-code tools just keep getting better and better? I think there’s a potential fear from those of us that are used to writing all of the code that we might be replaced by a low-code or no-code service.
[00:39:20] RC: Yeah. That was probably one of my fears at the beginning, but I feel like after kind of using these tools a little bit more, I don’t think that fear is so much justified anymore. I mean, someone still has to create those low-code tools. So there’s definitely engineers behind those stools themselves. I feel like even in our case, we’re using a low-code tool just to help us with the back end. We’re still doing stuff on the front end, and while yes there’s low-code tools that can help you on the front end too, eventually there is a level of customization that you would need specifically for your use case. I don’t think a low-code tool can really help you with some of your specific business logic, some of your specific things that are unique to your app or whatever. Even for us on the back end, there’s still some work we have to do to kind of make it work for us. So I think the low-code tools are doing more to like just kind of help developers focus on more interesting things, rather than like kind of the same stuff over and over of authentication database stuff and all that. We can just focus on the cooler features that go into our apps.
[00:40:27] SY: Do you think non-developers can make use of low-code tools? Or does it make more sense for developers who already know how to code?
[00:40:33] RC: I think non-developers could definitely make you some of them. On Twitter, I’ve been seeing a bunch of folks who’ve been taking different ideas and starting them off and using low-code tools, especially if you’re, I guess, a non-technical startup founder and you’re having trouble, I guess, finding someone to work with you on building whatever your idea is. It’s awesome to be able to get started on that prototype to validate your idea and then afterwards an engineer can come in and help you take it further than that.
[00:41:06] JP: How do you see these low-code tools scaling as your business grows? Do you think they are a good tool just to get started with? Or can you see your business using these low-code tools even after you scale, even after you grow and hire even more developers?
[00:41:25] RC: I think it depends. Like in our case, I feel like for the back-end services that we’re using, we’re using services that huge companies are out there using, like AWS Cognito Services, DynamoDB, and all that. We’re getting access to tools that are proven to scale. I would say depending on the tool you’re using, depending on what you’re building, yeah, there could be some scaling issues, but I think that’s where kind of like developers can come in later down the road to help you with those issues. So I think at first, probably yeah, low-code tools are probably super useful for getting started, getting those prototypes out there. But there would still be some level of when we’re scaling, what do we need to change? What do we need to do to make sure our users still have a good experience and all that? Which kind of, I guess, goes back to like why I think we don’t really need to worry about, I guess, low-code tools needing to replace us.
[00:42:20] SY: Well, thank you so much for joining us.
[00:42:22] RC: Happy to be here.
[00:42:34] SY: Thank you for listening to DevNews. This show is produced and mixed by Levi Sharpe. Editorial oversight is provided by Peter Frank, Ben Halpern, and Jess Lee. Our theme music is by Dan Powell. If you have any questions or comments, dial into our Google Voice at +1 (929) 500-1513 or email us at [email protected] Please rate and subscribe to this show on Apple Podcasts.