 Good morning. Good afternoon. Good evening and welcome to another edition of the developer experience office hours here on OpenShift TV. I am Chris Short, executive producer of OpenShift TV. I am joined by two of my favorite Red Haters, Serena from the Future and Ryan Jarvenin. Serena Nichols is her real name. Today Serena, we're talking about what exactly? Today we're going to give you an overview of our 2020 developer survey that we took around the OpenShift experience. Wonderful. Looking forward to sharing that with you guys. Yeah, this should be fun and enlightening at the very least. So do we want to kick it off right away? Yeah, I mean, if anybody has any like office hours type questions for us, feel free to ask. But yeah, until someone asks something, let's go ahead and fire away here. I'm going to head on over to the chat on Twitch. We also have a chat running through YouTube live via restream. So feel free to join us on either of those outlets. You can find all the links on OpenShift.TV. And yeah, I'll be keeping an eye on chat. So definitely let us know if you have new topics that you would like to see on this show. This is intended to be an office hour segment. So definitely get active, ask your questions. We'll do our best to respond to any that you have. And yeah, we love talking about Kubernetes and developments and how you can make the most of your development time while taking advantage of Kubernetes. Hopefully this developer survey has really kind of collected a lot of our feedback from you in the past in 2020. And we're anxious to hear what you've got on your mind going forward in 2021. Awesome. Carlos Santana, one of our regulars and fellow IBMer, not fellow, we're Red Headers. He's an IBMer, but you know, one team one fight kind of deal has a tecton trigger deployment question for you, Ryan. I wonder if you can grab that. I will take a look. Yeah. The question is how can I hide a tecton trigger deployment from the UI topology. It's very annoying to have the automatic deployment there is noise. Talking about like get ops add on type stuff. Yeah, yeah. I know we have some support for hiding things from the menus. I don't know if we have a filter that will easily allow you to hide things from the topology view at this point yet. That sounds like a good feature request though to be able to filter things from the topology as well as from the other lists. I think most of our filtering so far has been around kind of categories of things like D. It's probably not the exact example but like hiding. I don't know like if you didn't want to show certain operators or certain other things in the left hand column you can I think get some of those filtered out. Take a look at the console customization episodes of our developer experience office hours for more information about what console customizations are available today and what what you're capable of hiding and showing currently. Yeah, let me see if I can find that one. I'm loving this have no fear YouTube is here. It reminds me of an old cartoon from back in the day. Yep. Yeah. All right you want to dive in. I will happy to yeah definitely. Okay, so, as I mentioned, we're going to share some of the survey analysis of the 2020 survey. This is just showing that we did have shoe one was our data scientists associated with this and Steve spiker who's usually kind of co presenting on something like this is isn't available today so I'm just going to be talking about it today. You know the background is around to learn more about how developers are interacting with open shift and help drive product direction of course, this is the third year that we've run the survey. And the survey ran from October 4 to November 22. So about six weeks at the end of the year, and we had 38 questions. So I'm going to try to give a quick quick overview of where we landed and I can also play and I'll let me know if I need to pause for questions along the way. Most of the questions are on a reading of scale of one to four where one is lowest and considered very unsatisfied and high is is the number four and very satisfied. You'll also see three different bars for most of the top of most of the answers where green is representing 2018. And just 2019 and that hot pink color is representing 2020. And then you can also see on the bottom on the x axis there is a scale and that's kind of how we're indicating percentage across the board. Nice. The other thing is that we do, you know, around a metric of understanding the deviation or the possibility of what is it called. The answers might not be specifically correct. There's a chance that there's like an 8% right dip positive minus on each of these things based on the sample size that we have in some of the metrics that we had so just remember that as well. And so if we start looking into some of the questions we asked the first one was around what type of application are you running on open shift. This is pretty interesting where the micro microservices stateless apps, stateful apps and traditional apps are definitely the most popular but if you kind of look at where the numbers are trending from 2018 19 and 20, you can see a decrease in there. About three quarters of the way down or two thirds of the way down the page you can see that we did add event driven applications and so that was a new item that came in but it did come in at six so it's an interesting thing where we'll likely start to see that trend where that's going to take an upswing across the board. And then there were some other options on the bottom that continued to gain a little bit from previous years but nothing significant. What we also did was we kind of pulled in some of the slides around the CNCF survey to try to back up some of the data that we heard or got from our, our own survey. So this is around do you run stateful apps and containers and, according to the season CF survey 55% are yes running in production. Pretty interesting it also talks about 12% are evaluating and 11% plan to use them in the next 12 months. So, again, some similarities of the trend that we saw through our survey. And that is again from the CNCF how is your organization using serverless technology. And so just under a third of the respondents are using serverless technologies in production and another 20% 21% are evaluating and 14 plan to use in the next 12 months. Interesting information. As you know, you all know in the console which is where I'm spending a lot of my time we have a pretty invest a pretty large investment around the serverless experience on eventing and serving. And so I think this maps well into what we're seeing and reading here, right. So the question around how do you describe your serverless fuzz application. So data processing is the most popular type of those types of apps, and then you know the second third and fourth are all around the same percentage so around 50% of the people who answer this question said that they are describing apps and back end or REST APIs. So around the question of what did most people what what version are was the most recent that they had experience with you can see a good decline, a decent decline in the three dot 11 or earlier which is great right if you kind of focus on that 311 or earlier 80% of the people respondents in 2019 said 311. Now we've got about 3637% in 2020 so we're seeing that decline which is what we're expecting. The thing that is interesting about the survey is that we are getting respondents that have of people that are using three dot x as well as for so we're not quite. So some of the the answers that are we we're getting are obviously are also mixed based on three versus four so it's a little bit hard to kind of evaluate if some of those answers are more focused on one version or the other so we're hoping in 2021 when we do the survey again. That most of our customers will be off of the three dot x platform and or we were are also going to change the way that we're asking questions so that we get direction on the four dot x experience, even if three dot x is something that there's also continuing to use. I like how many people were up on the four dot five on the kind of later versions there was like a little gap between four dot five and four dot one. You know you've got the people who have just started looking at four and never ran their upgrades, and then people who did probably continue running upgrades and ended up on four five or maybe even four six, eventually so cool to see that people didn't just switch to four, and then stay put, you know the hopefully the auto upgrade, or the support for in place cluster upgrades is working well for folks so it's cool to see some of that shift in the data. Usually we have kind of a slow uptake on some of these releases so nice to see it moving along. Yeah, that's a great observation. Yeah, I agree with that. For sure. Okay, so kind of a summary on that section the top up five apps decreased overall, but they were still the top five and as we noted there was some growth in this in the next six through nine and eventing or I'm sorry serverless was one of those that was kind of looking possibly a major role in the future. The newest version tried in 2019 80% of the people said three dot X and like you just mentioned that had decreased in to 40% in 2020, but we're also seeing a high uptake in what we have for four dot X. And I apologize for trying to handle my dog here at the same time. We all have these problems now. Dogs, kids, whatever it is. Yeah, it doesn't matter right like Max just ran in here a couple minutes ago to give me something I was like, thanks kid. I wasn't doing anything. Okay, so as far as the actions like I mentioned we're thinking about how we can focus the survey or a word the survey a little bit more so that we make sure that as you're going to see in the in the next couple of sections we talk about ease of use and satisfaction we want to make sure that those ratings are based on four dot X in the future so that we So that's not mixed. So we have a better understanding of the baseline. So the next sections around developing and testing code on open shifts. So the first question this first question here is around is open shift easy to use rate your agreement with the statement. So here this is one of the outliers and we had a scale of one to five or strongly agrees somewhat agree. Kind of neutral, and then somewhat disagree and strongly disagree, but at an overall overall level respondents find you open using open shift relatively easy to use it's kind of a higher up on that top side. We're seeing a slight increase between 19 and 20 nothing substantial. You know, a couple of percentage points but nothing substantial there, but still it's good to see that we're on the higher end of that scale. And now this is an interesting one where we're showing rate your agreement on open shift is easy to use based on the release you are on and I probably cannot increase this percentage so I apologize for this. But what you do see here is in four or five we see a pretty good uptake on what people are doing for somewhat agree and agree and strongly agree on four dot five, where we have I think it's between the top two we've got about 25% are saying they either saw strongly or somewhat somewhat agree. But then we're seeing a little bit a big fall in when you look at four dot six. So we got to investigate that a little bit it could be that four dot six had just come out. Maybe people weren't used to it yet. Not sure what features they were trying to use. But that's something that we'll definitely have to take a look at. So some feedback on that last slide that the font is a little bit on the small side. Yeah, on this one. Do you know if we're planning on dropping a link to the deck at some point or or not today. I don't think that we're going to drop it today. I think we need to improve some of like the readability of the font. Yeah, makes sense. Yes. After after we get a chance to do some cleanup, we will share the results a little bit more more widely. Thanks for asking in chat. Definitely. Yeah. Thanks for that. Okay, so this one. So now we're going back to open shifted easy to use, which is interesting. So we've got the 311 or earlier is on the bottom on the bottom half of the page and four one to four six is on the top half of the page. And I think this is great. If you're looking at ease of use from four one to four six, we are seeing higher percentage right so 50% of the people are saying they somewhat agree, and about 17% are saying that they strongly agree so in, as opposed to in 311, the percentage was much lower I think for those talk to ratings. So that's, that's good news when you're talking about relative ease of use from one product from one release to the next. When we're talking about capabilities meeting our requirements, almost all the respondents found that open shifts capabilities do meet their requirements you can see again a little bit of an update up up tick in somewhat agreed from 2015 to 2020, but everything else is relatively similar across the board. The next section of the survey is really focused on satisfaction with developing and testing on open shift. So rather than showing each of the individual charts, I'm going to give some more summary type information so the questions that we talked about included the satisfaction around the ease of getting started interactive local development, open shift environment used to develop and test against the image build process and source to image, and the time it takes to build and run an application. So again would be interesting for people to continue to comment if you have. If you agree or or really even disagree on any of these it would be interesting to know right. So our observations was that the satisfaction of interactive local develop development made a big jump in 2020, and that had previously been one of our lowest rated items. The getting started piece kind of does remain flat so we're trying to figure out how much of that is attributed to be three versus four mixed experience. And there was also some interestingly high feedback with four one to four three, and then good with four five and then again a little bit negative with four six but there was a small sample size so we need to do some investigation there. And then the nice thing to see here was that with the steady improvements, we saw across the board from image build process in S2 I were in 2009. I'm sorry 2018 it was 19% to then to 23% then to 31% for that very satisfied reading. And as far as actions go, we were talking about the fact that it's going to be interesting to see a satisfaction regarding getting started will increase because we've got all these, you know, great improvements and other features that are available, starting recently with the developer sandbox, the, you know, making the footprint size smaller for CRC in the console specifically around getting started we've introduced all these new quick starts, some in four six but more and four seven and guided tours as well. So we're hoping that those things will help to improve some of these ratings in 2021. Cool. So we have enough choices around utilizing services on open shift so this kind of focus on satisfaction around being able to find a service, whether it's cloud based or local, having enough choices of service, being able to set them up, maintain them, as well as use them inside of your And again you can see some, you know, not super significant again like there's two or 3% I guess increase or decrease decrease on very satisfied, I'm sorry increase on the very satisfied rating. Decrease on the satisfied rating because I'm assuming that those people kind of voted up, which is good, good to see that's finding services, whether it's cloud based or local. But for the most part customer satisfaction is pretty flat over the last two years in this area. We're very minor improvements around having enough choice of services, maintaining services or user services in the applications. And so again I just mentioned some of the things here that we're working on in 4.8 in the console, where we have an enhanced catalog experience as well as new topology feature to help find services so we think that that will really bump that finding service piece. When you have the red hat integration camel K operator installed developers will also now see in the event source catalog camel K connectors, starting with a few you know starting I think with five, but you'll see that number increase so again the number of options that developers will have available to them will continue to increase. And then there's this, there's a bunch of work going on with manage Kafka services, which will set the stage for a consistent experience in the console around managed services across the board so I think that's super important as well. That's interesting to me that managed services didn't see a big increase with all the operator hub options that are potentially available. Not all of them are pre installed on your cluster but I was kind of expecting to see a big jump in availability of related services that are just easily pluggable and and instantly kind of available, given you have the admin capabilities to add the operator right. Yeah, I agree and this could still and this is the tough part with this one where we still had a number of people back on three dot x the question is exactly. So again, we'll definitely be addressing that that part of the survey next next time around so that we can have a better understanding. I'm also curious like there are lots of operators available like I wonder how cluster admins are picking and choosing which ones to take right like yeah. I'm not sure if we do much research around that or have much knowledge like, I think we, we blog on quite a few of them and what some of them will bring bring to the table but I'm not sure where, where people are getting. How they're making their choices. Yeah, JP data have updates specifically for you from our product management team on quays requirement for Docker 34 is getting released this week, and there will be improvements in the documentation to handle pod man versus Docker better. So, just wait till the end of the week and you'll have an answer hopefully. Okay. So application development features on open shift the section of the survey focused on application development but specifically around application health performance visibility feedback on progress and errors during development application debugging project and environment integration with CI CD, as well as availability of modern frameworks and runtime. So again just kind of skipping this charts and just moving down to our observations and active in our actions. Overall customer satisfaction is pretty flat again. In 2020 application debugging and feedback on progress and errors are the lowest rated items around satisfaction that we have. As far as actions go, you know, there are ongoing efforts to improve application health and performance visibility. We've got a lot, a lot of work. We have a monitoring slash observability team that's doing a lot of work on the back end and the console team is is working with them to see how we can bring pieces in. We've also been working with other members around, you know, Java performance is their way for us to bring in Java performance dashboards into the console, etc. As far as like the the one around availability of modern frameworks and runtimes, what we are doing in four dot eight is we're, you know, there's more and more corkis. The corkis experience is showing up more and we've added some quick start so that to provide some awareness for users that that stuff is available and can work on the open shift platform and how I'm kind of walk you through how to do that. And the other thing I did want to note too is since integration with CI CD tools was part of that, you know, four dot five through four dot eight is providing a seamless integration with tecton pipelines inside the console. We also have like a, there's an ID that's available. I'm sorry, a plugin that's available for for ID ease around tecton as well. So I think, again, we're providing additional pieces of the story that should continue to help those satisfaction rates ratings go up, I think. I particularly appreciate that last section on developer productivity. I like the focus on feedback during my development loop and that's something that's not always available depending on how you have set up your development environment. A lot of folks will do a very minimal setup where most of their work is being done, it may be locally in an IDE. They may not be leveraging their cluster to get that kind of production quality feedback as part of their local dev loop, they might only be getting that feedback after they deploy or push something into a tecton pipeline or others, right. So you get that feedback on an interval, but not necessarily as you're doing the iterative development work. So we've added a lot of capabilities to get more feedback out of your cluster production quality feedback, while you're actually doing that iterative work to hopefully give you just a better reflection of what production is like. And to help you save time and energy as you're working, leveraging that improved feedback. So I know we're doing a lot of work to help connect the dots for folks on those pieces, but a lot of new things in 2020 and more on the road coming up soon. Yeah, for sure. This piece is interesting as well so this is around resources for OpenShift and we're not talking about Kubernetes resources but more like resources for a user to go get more information right. It focused on satisfaction of resources for OpenShift around product help and documentation, blogs, whether they go, you know, blogs covering developers.redhat.com as well as the OpenShift blog. And I'm noting that here because we don't have a satisfaction rating based on each of the different areas it's on blogs overall, as well as how to guides and tutorials. So our numbers aren't terrific here by any means. There's a lot of improvement, I think room for improvement. And you can see like there's slight changes, but it's not, it's nothing really substantial across the board for product help blogs you can see we have time from the very satisfied rating we've gone up from 2018. What is that maybe. You can say it went from about 18% to 30%. So that's pretty decent. But again, we really want to be I think a little bit higher than that, if we can be. And then around how to guides and tutorials. This seems even a little bit lower. So that being said, I think we do have kind of some good news in the pipelines as well though where we're going so again the observations are that product help docs, how to guides, how to guides and tutorials are some of the lowest rated areas across the board. We did see some slippage in satisfaction in product help and docs and blogs from 2019 to 2020. And we clearly need to focus in these areas but we are anticipated anticipating specifically in the console that quick starts are going to address some of these areas around the tutorial piece. And it's really interesting so we are utilizing quick starts are trying to utilize quick starts from the PM and the UX side but we're also talking the developer advocate team on like how they might be able to take advantage of them. We've also had a number of meetings with some customers who are really interested in bringing quick starts into their own environment and utilizing those to help to teach best practices on how to create or, you know, create upgrade maintain applications in their own environment for their own developers. So that's pretty interesting to see, I think. And I guess that last bullet is we do have a plan on focusing focused on how we're improving blog content delivery as well as contribution. I'd really be interested if people have the opportunity to kind of set say something in the chat like what are the most useful types of blogs that you guys see. Do you like the what's new blogs do you like more something around it's walking you through something and you can learn about something really specific. And or are you are you typically looking at open shift blog versus the red hat. You know the rhd blog site. So, there's a bunch of questions there but if you want to provide any feedback you know any of that would be much appreciated. Any questions yeah we definitely yeah if you have ideas or feedback from, you know, what you want to see on the blog let us know for sure. Yeah, what's new is a blurb marketing, not good should point to a video tutorial like the ones from Daniel. I mean we do have, and we are trying there is a what's new page now. If you go to open shift comm slash what's new, you'll get the latest briefing and deck from the last you know show that we did on the topic. Interesting JP date says I like who I look for who they're from. That's an interesting one. Yeah, anyone have a favorite post. I mean, there's so many now. Andy has his hands full our blog curator as it were. Yeah, I haven't been as active on the blog personally, but I do see posts occasionally from our product management team that I think are very informative and have a lot of really great details on roadmap and and new releases and other things like that so. Yeah, definitely let us know in the chat if you have a favorite blog post or another blog that you think has vastly superior content that I should be trying to take notes on whatever they're doing and work some of that content into our blog. Definitely let us know we value your feedback. Yeah, and you know good feedback is the challenges you're hitting right like GP date was talking about his challenges with installing quake quake key however you want to say it on real late. Right, like we need feedback like that to get you all great blog post to help you enable yourselves to go hard and fast right like that's the idea move move quickly and safely. I'm also just going to pull these in here I'm just bringing out a couple of tabs so I think probably everybody knows about the open shift calm blog I'm sure. One of the things that we have been doing is a concerted effort when we're doing something for the developer is that we are posting our blogs on developers dot read at calm. I'm not sure if you guys are all signed up for that program or not. Definitely look into it if you are are interested but if you go into the products tab on the top right. What did I just do and then scroll down to the bottom there's a red hat open shift here, and this is kind of where you can see downloads getting started and overview. But this is where you can see a lot of the content that we do. I think it's underneath getting started as well. There's a bunch of the resources that we have around the developer perspective and developer console here as well as other areas around developing on open shift so definitely take a look at that as well. Have Twitter notifications for open shift account. That's that's a really good tip. Yeah. Yeah. So is there are people looking at just like all the blogs. This is a good question that Ryan brings up. You know are you looking at just the open shift blog are you looking at the developers blog because the content does kind of cross pollinate seldomly I feel like right like we keep one in one place one of the other kind of deal. Our survey did not differentiate just so you know. Okay. And it's actually I mean it's also pretty. Yeah exactly. And it's also pretty interesting too right like so we you think about like what we're doing with their developer experience office hour we oftentimes are talking about how to customize the experience for developers, which might be tailored to admins right that the admins have to do the customization whether they expose operators or shut off helm or add more helm repositories or whatever. But it's they're still an overlap so it is it is kind of interesting to see where people kind of innately go for for those blogs. But I'm going to continue on a little bit. So this is around satisfaction with open shift build features. So this was reported just a little bit differently if you remember though one was on a scale of one to four one is very unsatisfied for is very satisfied. So we're seeing numbers kind of right around satisfied I guess. So pushing builds build artifacts my repo is three one along with security concerns with builds and ability to apply runtime config and availability availability of builder images. And then the straight three dot oh around injecting dependencies and ability the ability to extend and modify build steps, and then a little bit lower around build speed and the resulting image size. So some, you know, observations and summary around that is the overall sack satisfaction. Again, it's pretty much everything is relatively flat this in this instance slightly up. Unless we just noted dependency injection and image size and speed build speed or rank the lowest. So part of those actions are to prioritize some of those features in the upcoming roadmap and investigate dependency injection issues. So, and also hoping again this goes back to guidance and documentation best practices on how to achieve those kind of build improvements. What are the biggest challenges you face developing applications with open shift. Some of these works are going to start with the CNCF survey where what their challenges were in using and deploying containers. So their top challenges were complexity joined cultural changes with the development team. So that is kind of interesting. And then if you take a look at some of the our things on our survey, the observations were top issues were time and culture, local environment and CNC new and getting started or documentation performance and resources and stability. So it is interesting and that that new user getting started documentation and performances performance or being able to kind of observe or, you know, monitor things are seen to be some of our largest challenges there. So, again, around actions, increased priority of features and documentation to reduce friction for adoption, and of course, continue to try to reduce that CRC footprint which is already in the plan and we've definitely made some improvements to date but we're continuing to work towards that. I like, I noticed that last slide had was was more general. Yeah, CNCF survey, and I think this noting that complexity and cultural changes with development where some of the top kind of hold ups to to getting top challenges is kind of relates to some of the issues we saw earlier with getting started being like a big challenge. We had some kind of lower scores around the getting started experience and I think part of that is just for folks who are moving to Kubernetes or two containers from a before Kubernetes type of era. There is a lot of it's a there's a lot of knowledge, a lot of learning, a lot of skills you have to adopt, and all of that takes time all of that takes energy. Hopefully open shift is doing a good job of hoping of helping reduce the complexity and the amount of learning that you have to do. It's definitely a challenging situation because we want to give you full access to all the underlying Kubernetes resources, while still hiding enough of it to make it easy for new users. So, hopefully we're doing a decent job of kind of splitting that balance there. I mean, it's always hard when you're dealing with abstractions, right? Like, yeah, how much do you abstract away? How much do you expose that's always like a razor edge balance, I feel like. Yeah. I'm just noticing something here that somebody had mentioned it would be nice to get if there was a open shift only newsletter I'd subscribe to get it. If it was once a week, but I'm not going to visit the sites every week to check what's new. So I don't know if anybody who's who are like Chris or Ryan, I don't, I do not know this, but the red hat developers newsletter. Do we know how frequently that goes out. Is that a month? I guess once a week. I forget. It might be every other week. I forget to be honest with you. Yeah, something to check out. I mean, I can take that action so come next week we can talk about that a little bit too. Yeah, I'm probably going to reach out to Alex Handy to just be like, it might be a good idea to start a newsletter with just blog posts and interesting things, you know, like links and summaries done, right? Right. Yeah, five to 10 links a week. I feel like that would be a good thing. Definitely. Definitely. Okay, so let's go. Getting started in languages in framework. So how do you get started creating new applications for open shift. So I will read some of these to you because I know that the font is very small. So the first section is around the web console. The next thing that we're seeing a decline on that I know these are images to so I can't do anything unfortunately. The next one is around OC or cube cuddle creates. The next one is around internally develop internally developed tooling, and then helm charts is the fourth open shift templates is the one that's only highlighted in pink. Audio create. So that's kind of where we are with the top ones. So the kind of overview is that many respondents start creating new applications on open shift by using the console or or the OC or cube cuddle create commands. Interestingly enough, when I do have conversations with people with customers, what we are hearing is on board, you know, brand people who are brand new to open shift are typically coming in and doing like the import from get that type of thing with the console, and are very familiar and comfortable with that, because they're the ones that are not super comfortable with Kubernetes and want that more abstracted view. Again, interesting to see what we're what we're showing here. It looks like helm finally moved ahead of internal internally developed tooling got surpassed by helm. Right, right, right. That's cool. That's cool to see folks move to standard kit that is maintained by a community rather than all in house. There's definitely some value to doing it all in house when you need to, but it's nice to see helm charts kind of moving up as a widely adopted way of packaging and distributing your solutions so that's that's pretty cool. Yeah, the interesting thing like I'm just trying to look at the percentage so in 2019 home charts was it about 18%. And now is it about 37%. So that's a pretty big hike. Yeah. Yeah. Okay, the next one again is around which languages and tools are used by your applications on open shift so I'll take a read down like the top eight or so so Java was number one open JDK to JavaScript. Python, jboss EAP, go Tomcat, jboss web server, and then dot net core. And then we also see corkis. Again, corkis is new so you see corkis kind of where they they were at about 12% last year and then this year at about 24%. So it will be interesting to see if that continues to grow or it will be interesting to watch that grow. And it's also well interesting that the top languages used by apps and open shift or Java with JDK JavaScript and Python but you're also seeing a decline from 2018 to 20 with a few of these. So people disagree and think go is go is the king here. You know, I used to see charts like this where they would have JavaScript and NodeJS as two separate categories. It looks like we do. What's that? I mean, is that really like a thing? I mean, are they really different that much? They're both JavaScript ones just running on the server. So it could be like static web versus actually run via a NodeJS server that you want to distinguish between which is somewhat valid. But yeah, it's interesting to me that there's a Java and then an open JDK and then also wild fly and like a couple potential ways to roll up Java into a larger category. But since this is Red Hat, Red Hat Survey, we do have a lot of Java adoption. So it was kind of expecting to see Java up at the top of the list for sure. Totally. Wasm for next year's survey as Carlos Santana. I mean, that's probably going to be in the mix at some point, right? Yeah. I heard last week some folks from the Rust community saying that Rust is potentially the future of JavaScript because you could write Rust that'll compile to binary in your browser. And they're like, we're just going to skip over everything and go compile right to binary and ship it to the browser. I don't know. That could very well be the case, right? Like Rust has a very bright future when it comes to WebAssembly. Yeah. All right. So this next slide is around, like we pulled in a part of the GitHub survey as well, right? Just to show top languages over the years as well. So if you can't read the image on the left, 2014, it starts on the left-hand side and moves over to 2020 on the right-hand side. So you can see JavaScript remaining consistent. You see Python kind of with an upswing from 2014. Java maybe a little bit of a downswing. Yeah. Like I think it's interesting that Python has kind of grown over the years, right? Like remember like 10 years ago, people were like, oh, that's like a teaching language. And now it's like, nope, it's everything language. There's really good AI and really good libraries for machine learning, image processing. I know there's great libraries for a lot of different languages. A lot of stuff, yeah. Python has a great collection of libraries that are available. Yeah. TypeScript is making major moves there in that last one. Yeah, here it is definitely. Isn't TypeScript just like safe JavaScript, right? Yeah, yeah. Usually it's kind of a new fangled JavaScript to some extent, yeah. And then here is another survey, Stack Overflow, kind of the same, you know, trying to tease out the same types of information, but where they see Python moving down for eighth year in a row, JavaScript is maintained at strongholds, you know, so kind of some interesting things. And we're seeing some similarities between these, so it's interesting. It's amazing to see Ruby just drop so far so fast. Yeah. HTML. Okay, then. So which frameworks are you used by your applications in OpenShift? So the first. So again, I'll read down the left axis is spring and spring boot is the first one. Then the next one is Node.js. Then the next one is Kafka, then followed by Angular, React, Hibernate, Camel, and Django. So Kafka was added in 2020 and is already in a top three, which is pretty interesting. So about, let's see, just about 44% of the respondents said that they're using Kafka. So according, again, based on our survey, the most popular frameworks are Spring, Spring Boot, Node.js, Kafka, and Angular. Interesting to see Angular on that list. Yeah. That's more front end, right? Or am I just off the base on that one? Angular and React, right? React is definitely a front end framework, I feel like. Yeah. So it's interesting. Let's see, the next one is how do you migrate existing applications to OpenShift? So lift and shift, copy existing apps, resources, and modify them using OCMU app with existing Git repo. Fourth is refactor and rehost. Fifth was third party consulting. Then strangler pattern, replacing monoliths, capabilities with independent microservices. Then internally developed tooling. Then Red Hat migration tooling for applications and others. So it is interesting. I'm just going to note that really quickly. Red Hat migration toolkit for applications was newly added in 2020 and about 17% of the people have noted that, which is interesting. So it will be cool to see if that continues to grow as well. Go ahead. I feel like we've done several migration toolkit shows, and I believe there's one upcoming too. Great. So let me grab some links for you all. Yeah, that would be great. Please keep going. Yeah. Okay, dope. Yeah, so clearly lift and shift and copy existing application resources and modifier are the top two. But we have seen an increase with lift and shift between 20, between the last two years in 2020. So that's interesting to see. I'm curious to see, I'm curious to see if we have an operator hosted pattern showing up here in future results. I know that's something we talk about being a way of hosting stateful applications because you could. If you had a step where you needed to bootstrap a database and pre populate it with a set of data or do backups or do other more complicated app performance things in the background. Operators are a big way of achieving that goal, but I don't see them listed here yet. I don't know how widely adopted that is as packaging up your app in an operator. But I would hope to see more adoption of that, especially with our support for operator backed helm charts and other solutions like that. That's really interesting. So Ryan, we should definitely chat because making sure we don't have that as an option today on this. So we can chat to see if we can get that out over 2021 for sure. I wouldn't expect to see a huge amount of it currently based on the current documentation around operators most of it's around how to package up your your database solution, more than your application but definitely an option for folks in the audience. So packaging an app in an operator is twice the complexity a client would not go with that route by experience. Interesting. Okay. Yeah, it's I mean it's basically asking someone to package their application by writing another application in in go. Right. And not everyone is is up for that challenge of like oh go write a second app in go to package your first it's it's a yeah it's not super easy, but with support for helm charts, operator backed helm charts. So that might be a little bit easier option, but there's definitely some trade offs and more work for us to document how you might adopt one of those approaches. Yeah, and there's, there's a lot of ways to write operators now. Python bash ansible. The list just goes on and on I saw another project I'm pretty sure a Java based operator thing was out there I think I saw last year. Yeah. I'm going through some of the observations audio I'm not sure how many people are on the call are using that but we, we, it made a jump from a percentage perspective from five to 13%. So that's great news but the overall usage is still low I think we we need to continue to kind of make people aware of that and the power, explain the power and why somebody would want to use that CLI. And as helm charts, helm chart usage for apps continues to increase which is great. We're and as everybody knows we are increasing the amount of support inside of the console as well as some other areas. Over 50% of the respondents are using the console to start application creation. We're going to play the next couple of items. Some other on the action side. We are starting to gather metrics for audio usage, as well as some of our other areas in the portfolio. Helm certification process is going to be started and will be supported in the console sometime within 2021, which is, I think good news. We need to improve enhance the console and support multiple methods of creation dev files or something that's really exciting that people are talking about which I, you know that's going to help satisfy our kind of roar launcher use cases. And those are going to be, you know, I think those dev files are pretty strong and flexible. And I think that's going to be a really good option in the future, as you'll also see our portfolio kind of standardizing around the usage of dev files 2.0 as well. So looking at the clock, I'll just do this one section. So where do you develop and test your application for OpenShift. So the first one is around local desktop instead of OpenShift. Other internally hosted local on desktop with application runtime externally hosted or local desktop instance of other container platforms. So a large number of respondents use a local instance of OpenShift or internally hosted versions for developing and testing their apps, which is interesting. You can see though it did kind of seem to increase, I'm sorry, decrease a bit on the local desktop instance of OpenShift from 2018 till today. So what do you use to run a desktop instance of OpenShift. So we have code ready containers is the first one, the first option mini shift is the second option OC cluster up is third container development CDKs for and then other. So we do see again, a big kind of uptick on code ready containers where in 2019 it was at about 39% and has gone to 60, 64% or so, which is exciting. And we do see mini shift declining. So, as well as OC cluster up. CRC really are these people rich with large laptops. It's entirely possible that they're running CRC and VMs in their environment right like we don't necessarily know the underlying infrastructure CRC do we, we don't get telemetry on that right. I did see some data on most installed CRC versus installed to AWS or installed to Google's cloud or and there was a surprising amount of adoption on CRC so you're right I don't know whether that is 100% of that is going on to laptops. You could definitely run it in a large VM hosted somewhere. So yeah that's totally a valid valid way of setting this up. And so this one is around which local desktop instance of other container platforms do you use in Docker run. So, again Docker runs first mini cube second pod man run third other and then Docker swarm so again just interesting interesting information. And this is aligned. Now we pull back in the CNCF survey, where the top Kubernetes environments are mini cube 37% for them on prep Kubernetes installations which is 31% and Docker Kubernetes which is 29%. So, for comparison, interesting. I think I'll just do this one last one, which editors do you use. And with this we're seeing BS code is number one. We do see an increase in fine. Yeah, yeah, we, yep. So VS code first vi second eclipse ID or red hat container ready studio, IntelliJ and JetBrains ID user fourth them as this. And then we get to visual studio code ready workspaces eclipse che and on and on underneath. But yes code vi and eclipse ID are the most popular editors for us. And so that's kind of where we stand in this more there but I think we're three minutes before the hour so I think we probably made made it through quite a bit of this. Any questions or comments around what we see today. No, but there's like a lot of great chat. And I will make sure that you have all that feedback. After I get off here today how about that. My favorite comment from this page is is a these are all just YAML editors now. It's all YAML editors. I am a calendar driven YAML engineer. Still one of my favorite phrases yeah. All right, well thanks very much. Hey Ryan do we happen to have a link to that server that we put in last week. Yeah, let me see if I can find that real quick in our last minute or two. I had a link last time let's see. Yeah, so last week we had this little survey around you know what would you like to see in the developer experience office hours going forward. We kind of, Ryan did a great job at like notating some of the different types of options that we can provide and really getting feedback from you all to make sure that we're sharing what you want to hear so you can find that quickly. Yeah, trying to get it right now let's see. Actually, I think I might have tweeted it somewhere along the way. Let's see send this. The link with a short URL. Got it. All right. Perfect. Thank you. Yeah, pasted that into chat. And I should get a shorter URL for folks tuning in later, but yeah, there's a somewhat shortened URL. We would love to hear your ideas for additional topics on the show. There's a fill in the blank area at the bottom of that feedback form so definitely let us know if you have more topics you would like to see. And thank you to everyone who contributed in chat today. You've made it fun and engaging for me and I definitely appreciate it. So thanks again. Thanks to Serena and Steve Spiker. Did he help out with some of this? Yeah, thanks to both of you all on this excellent survey data. Really, really cool stuff in there. No problem. Thank you guys. Yeah, thank you all for tuning in up next. We're going to be talking about Stack Rocks, our most recent acquisition. So this will be a fun stream, but I've got to cut over to that. So thank you all for tuning in. We will see y'all next week or in the next minute or two. Take your pick. Thanks everybody. Have a great week.