The awkward state of the remote vs. in-person work debate

Even though many countries have ended the stringent lockdown orders that forced so many employees to go virtual, the conversation around remote vs. in-person work seems to have forever changed. Alongside that, looming threats and concerns stemming from COVID-19 continue to affect whether IT conferences such as Red Hat Summit, Oracle CloudWorld and others return to in-person events or stick with the all-online approach.

These factors, along with the continued proliferation of SaaS offerings, leaves many thinking that organizations will only shift more toward remote opportunities, rather than refocus on returning to the office.

TechTarget senior news writer Beth Pariseau shares these sentiments regarding the remote vs. in-person work debate, which she expands on with Ryan Black and Tim Culverhouse in this episode of Test & Release. Our conversation with Pariseau, who is an award-winning IT reporter with over 15 years of experience covering software technologies, touched on a variety of specific aspects within this debate, including:

• The current paradigm surrounding distributed, virtual and hybrid work approaches;
• Whether IT conferences regain any semblance of what they were like pre-pandemic;
• The vendor-specific concerns DevOps and cloud adopters should be aware of;
• The effect on digital transformation efforts amid remote vs in-person trade-offs; and
• The hidden security impacts that should be top-of-mind for IT.

Give this episode a listen -- or check out the transcript below -- for our discussion with Pariseau. Bookmark her author page and follow her on Twitter @PariseauTT for continuous news coverage on a variety of IT topics.

TRANSCRIPT

Ryan Black: Now, Beth, can you tell our listeners a little bit about yourself and what you generally cover for TechTarget?

Beth Pariseau: I'm a senior news writer -- with SearchIToperations.com primarily. But I also do some writing for SearchSoftwareQuality. I have been with TechTarget since prehistoric times in internet years -- since 2005. And I have covered everything from data storage to server virtualization to cloud and now DevOps and digital transformation.

Black: So, Beth, you attend a lot of conferences for your job. And we were very much curious about what you kind of sense the paradigm to be around distributed work, virtual work and hybrid work. I know that I think you're just getting back from attending a few conferences recently, yourself.

Pariseau: I'd say whether it's remote and hybrid work, or in person conferences, things are at an awkward stage right now. And it's not just my opinion, I also talked with the CEO of Red Hat, Paul Cormier, at Red Hat Summit in person last month, which was the first in person event that I've done since before the pandemic. And he was saying, and obviously, this is going to affect his business pretty deeply. But he [said] nobody really knows how this is going to shake out in another couple of years.

Right now, there's a little bit of a push and pull between kind of the supply side and the buy side in terms of employers and employees about what they're going to do -- are they're going to get back to the office, is there going to be a mandate to do that, are they going to continue with a kind of limbo, where it's hybrid or it's a mix?

But we have seen some pretty big companies and pretty big CEOs rebuffed in their efforts to push everyone back into the office. JPMorgan Chase is an example. I think companies right now have enough trouble retaining and recruiting new employees, given the way that the job market is with the Great Resignation and low unemployment, that they really can't use the stick to get people -- to just force people back into the office, people are just not going along with that, in part because of the way the job market is.

But also, in part because I think the perspective that the pandemic gave a lot of people, both that remote work can work demonstrably. There are some cases in which people are more productive working remotely, but also that that they're not necessarily going to be cowed by corporate culture as much as they were. Life is too short. So, when we all started working remotely, because the pandemic had struck, there was a critical impetus, and there was a definite direction . . . everybody had to go to their separate corners, because there was a life-or-death issue at hand.

And since then, there's sort of been, especially since vaccines came out and different people have different attitudes toward the risk of the virus, there's just been sort of entropy. And there isn't a forcing function to kind of push all that toothpaste back in the tube, so to speak. In some ways, you can't really undo what's been done. And there are some people who have had a taste of remote work that aren't going back.

Unfortunately, there isn't a definite answer yet. If the job market shifts and employers have more leverage, they may succeed in more of a mandate to get people back to the office. Or things will continue toward the remote and hybrid direction now that we've seen that it can work for people, we have the technology to make it happen effectively. But there's also a set of sectors of our economy that are tied to people going to work in offices, most especially corporate and commercial real estate and so who knows how those factors could affect everyone. Right now, if even Red Hat's CEO doesn't know, I certainly don't know how this is going to play out.

Tim Culverhouse: And you mentioned Beth, in terms of the return to office element vs. working from home or distributed teams: As part of your job, you go to conferences, you're reporting, you're doing live interviews. And we're starting to see the return of in-person conferences, but still with a potential either for online viewing or hybrid viewing.

And I'm just wondering, now that it's getting back to -- not a level of pre-pandemic, obviously -- but some semblance of travel with mass groups of people at conferences. What's your take from having been to a couple in-person ones, where there are large groups of people, and then also just the overall conference experience between sitting on your computer and alternating with Zoom rooms, and then going to in-person sessions once again?

Pariseau: I've done two events in person this year, for the first time since the pandemic: Red Hat Summit in Boston and cdCon in Austin, Texas. Both of them, I hadn't been to a cdCon in person, but I have been to a number of Red Hat Summits in the past. And in terms of the size, Red Hat Summit was not what it used to be. There were, we're talking in terms of hundreds of people there in person, versus thousands.

It had been held several times within the Boston Convention and Exhibition Center, which is [an] enormous, sprawling structure. And this time, it was held in the Westin [hotel] adjacent to that convention center. And so, you know, things are definitely not where they were, they kind of can't be. I thought cdCon was a similar thing. It had been a colocated event at KubeCon, it hadn't been its own multi-day conference before. So, I'm not sure how big it might have been, had it not been for COVID. But it was also fairly modestly sized.

I think both were good in terms of requiring proof of vaccination and precautions to be taken while people were sharing the space. I was glad to see that. And I thought cdCon was very energetic. The people who were there were engaged; they had big companies that I like to talk to, like Fidelity Investments, presenting -- who were also available to speak to. But the days of being at just a big teeming show with [and] going fishing on the show floor for contacts and for networking, which is really the big benefit for me of in person conferences, has not returned. I've made some contacts and I do think it's important to get back to in person for my job. It was worth the trip, it was worth the plane trip, to go to cdCon. But it's not what it was.

And then virtual conferences, it's not impossible to connect with people there. A lot of conferences -- especially in the open source world -- have Slack channels, which are really good. Also, LinkedIn. I tend to use LinkedIn even when I'm in person. These days, if I try to ask somebody for a business card, they look at me like, "You mean a piece of paper?" Nobody carries business cards anymore. So often connecting on LinkedIn is the way that you kind of capture a person that you're networking with.

Virtual conferences have their own stressors because when I'm at a physical conference, I'm not expected to keep up with every other news item that's going on in the market. But when I'm at home, it's a different expectation for whatever reason. And there have been times that I've done two -- and at one point, even three -- virtual events at once in the same week, which is also certainly not something I would have done in the physical world. So, they have their own pros and cons.

And the hybrid events are good. I was able to attend KubeCon Europe for the last couple of years because they were virtual and this year hybrid. I also heard about a cluster of COVID cases after KubeCon EU -- [the virus] is really not gone. It's really not completely over. And so, I think things are still sort of happening in fits and starts right now.

Black: To circle back -- and you touched on this, so I want to cover it a little bit: What from Red Hat [Summit] this year have you seen that DevOps practitioners specifically should have on their radar?

Pariseau: When I talked to Paul Cormier, he said very definitively -- and I hadn't seen him say this as definitively anywhere else -- that they're going SaaS-first with their products, including OpenShift. And I think you've seen that with Ansible as well. [Red Hat] did a deal with Microsoft Azure, where they're available with joint support on Azure in their marketplace, and people can put their committed spend in the Azure Cloud toward Ansible. So, they will, in some cases, be working with partners on that. And other cases, as with OpenShift online and their OpenShift cloud services, they'll be offering it themselves.

They are really pushing into not just cloud and hosted applications, but software as a service  managed services. Their OpenShift data science was an example that he gave of a service that they started software as a service and then went on prem[ises] when there was demand. So, that is a pretty big strategic shift for Red Hat to kind of go that definitively that that is their priority . . . and that's the business model that they want to pursue.

Culverhouse: And in terms of your discussion with the folks from Red Hat, and just out of curiosity: Did they say this, or is there an inference here, that possibly this approach to SaaS and the further push to the cloud has anything to do with the expanded time working remotely? Or is it just a better business model because they can find different ways to increase revenues?

Pariseau: I think it's more the latter . . . Every company, I mean -- name them -- they are trying to go to recurring subscription revenue, they're trying to go cloud-based, they're trying to go cloud-first. It's true of Splunk ServiceNow, VMware. They're all trying to get into this cloud-based and SaaS-first sort of model. I think that's primarily the motivator.

But the pandemic, the shift to remote work has given digital transformation and cloud migration a huge kick in the pants. What Cormier said to me was, "We're five years ahead of where we might have been without it." Because again, dire necessity forcing function. And it's either go cloud -- to kind of pull together these disparate people or resources in different locations -- or die. It was really like everybody really had to commit in that direction, where maybe they hadn't before. And the people who were best off, especially in the initial year [or] 18 months, were the people who already had started.

So, I think that's also been something that's been accelerated a lot by the pandemic, the shift to remote work, the need for things like identity and access management that isn't reliant on a VPN sort of cloud-native services that can scale. And maybe manage more managed services where you can't have your IT people in an office managing things like help desk from a central location for your company. So that tends to sort of move people toward the SaaS offerings. So, it's kind of a combo, but I think that shift had started without the pandemic.

Culverhouse: And as you said, just accelerated or forced the hand of a lot of these companies when they were inching toward it -- and now all of a sudden, it's full steam ahead.

Pariseau: Hugely. Yeah, the more you had not just gone to cloud, but to DevOps, Agile practices -- whether technical or organizational -- the better off you were, when the pandemic hit. The more able you were to work in a distributed team effectively; the more able you were to collaborate asynchronously, which is hugely important when people are remote and working from home. The more you are able to use these sorts of cloud services that have centralized resources, where you don't have to think about how to make them available for people, you just have to think about what you're doing with them. I think the further along you were in digital transformation, it's pretty obvious, the better off you've been.

Culverhouse: Has there been any shift or further awareness or even preventative maintenance for organizations that now almost exclusively work in these distributed cloud environments that when something goes wrong with your cloud provider, and everybody's remote, what's going to happen?

Pariseau: I put the Atlassian outage in a different category than AWS or any of the major hyperscalers having an outage, in part because the Atlassian outage in this case was much lengthier and more severe than any cloud hyperscaler outage I've heard of.

At least for a long, long time, I haven't heard of any cloud hyperscaler outage causing data loss that that the user wasn't at least in part responsible for. Those cloud companies like AWS are operating at a lower level of the stack. So, they're offering you infrastructure. But what you do with it, how resilient you make it, is up to you. It's up to you to have multi-region or multi-zone resiliency for your apps.

Maybe some people believe that, you know, multi-cloud is important for that reason. And it's up to you to set up the infrastructure to make that happen. Also, if you go from AWS to Azure to Google, cloud virtual instance or container cluster, is what it is, in all of those cases. I mean, maybe there's a little bit different flavorings to them, but it's still basically a server or cluster of servers that you're getting.

When you're talking about a SaaS provider like Atlassian, there isn't an equivalent to Jira. Or, if there is, it doesn't have your data, it doesn't have your history, it doesn't have your ongoing workflows that you're trying to do with it if you go to a competitor. And nobody's going to maintain dual workflows and different SaaS providers like one in GitLab and one, you know, Jira cloud for just such an occasion, which is still probably going to be vanishingly rare. Even if you're using something like Confluence as a wiki, it's not like Google Docs is a backup, it's not like it's an alternative . . . There isn't a way for you to shore up the resiliency.

Some people predicted that it might finally give SaaS backups the boost that they need. Although, having covered backups back in 2005, I don't know that it'll ever be a first-class citizen in IT, unfortunately. But there is the sense that at least you could protect your data from being lost. And I think that ultimately, there were not very many Atlassian customers that had anything like permanent data loss, but . . . it took some doing for them to get the data back that had been deleted from primary systems. So that's another difference.

I think this outage, the kind of cascading combination of problems that put them in this multi-week kind of nightmare, having to restore things from backups manually, and not being sure when they were going to get it back or whether they were going to get absolutely every byte of data back. That's not something you necessarily hear about these days very often in terms of cloud outages.

The argument usually there in terms of cloud migration is, "Is your IT department necessarily any better?" I don't think there's any enterprise that hasn't experienced that kind of outage, and maybe even data loss internally, with their own kind of IT team limited by size and number of people and resources to work with. And the other thing is that a cloud service provider has to pay you, or at least not charge you under your service level agreement. Whereas if your own personnel have some kind of outage like that, you still have to pay them. So, there are arguments for still kind of trusting cloud SaaS providers. But it does raise questions, absolutely, that such a thing could happen in 2022.

And people have been critical of Atlassian communication while it was happening. The other thing is, it's also very clear that Atlassian company strategy is those cloud products, and they are really pushing people to go there. And there have also been some pretty critical security vulnerabilities that have happened in their on-prem products that didn't affect the cloud. In particular, one that affected Confluence server and data center in the last couple of weeks, that did not affect their cloud instances of Confluence.

So, I mean, it's sort of a rock and a hard place, because you're not necessarily going to do better internally. You might feel like it, it's sort of like, you know, flying versus driving, you know, you're more likely to get in a car accident when you're driving yourself, but you feel that control, right? And also, because the scope of what can happen if something goes wrong in an airplane is pretty daunting, as in this case with this, this Atlassian outage. So, I had some analysts speculate that, that it might slow Atlassian's push to the cloud a little bit, they might rethink that strategy. I haven't seen any evidence that they're doing that. And I don't think it's going to stop the cloud juggernaut. I think people are going to figure out ways to maybe shore things up -- or just figure that the risk is generally lower when they go with a cloud provider than if they tried to do things themselves.

Black: I'd be curious. What thus far this year have you ended up covering that you in January 2022 would not have expected to report on?

Pariseau: That Atlassian outage, for sure. I definitely did not see that one coming. Nobody did. And even when it happened, at least at first, it was sort of ‘Okay, outages happen, they'll be back later today or tomorrow.' And then as it kind of went on, you started hearing reports of people not being able to communicate. Because unfortunately, one of the things that got lost in the initial snafu was the contact data for the accounts that were affected. So, they didn't necessarily notify everyone who was affected right away, because they didn't have the list right away. And also, people whose accounts were deleted couldn't actually access their service and support accounts, because they didn't have a valid account ID according to the system.

So, things got pretty ugly. And it took weeks for some people to get things restored. And it took days for some people to even be contacted, be communicated with about what was happening and what was being done to fix it. Atlassian also kind of held back external communications until they were more certain about what's going on. And I don't know, I think hindsight is 20/20. And you can say, "Oh, they should have communicated earlier." And in a postmortem, they said, they feel they should have communicated earlier. But if your communication is, "We don't know what the problem is yet, we don't know what we're going to do about it" -- I don't know if that's necessarily actually realistically going to help anything.

It was compounding errors that it reminded me a little bit of the WhatsApp outage that they had, because both were related to automated systems and scripts. And with great automation power can sometimes come great mistakes, and things can kind of get out of hand and cascade before you even really know it. I guess it shouldn't have been a surprise, given how interconnected, automated and distributed everything is, and the possibility for something to kind of be a domino effect. But maybe it's just a measure of how well these things generally perform that it was a real surprise that something this severe happened.

And, yeah, we'll see. I mean, they've taken some heat for that, they've taken some heat about the security of products like Confluence lately. And at this time last year, I think they just were still kind of viewed as a powerhouse and kind of a paragon within the DevOps and Agile world. And I don't think I expected them to have such a rough go of it. I don't think they did, either.

Black: What are your observations when it comes to acquisitions and buyouts right now in the industry?

Pariseau: I had expected there to be more. I think maybe everybody got too used to the heady days of last year when cash was sloshing around, and both venture capital and MA and in the market. And there's sort of been a correction. Also, inflation [is] playing a role in the economy in general. Ukraine, I mean, there's so much going on in the world, that's making things really volatile.

I think I had expected there to be more, I guess you would call it, a creative M&A, more where a bigger company wants to acquire a startup's innovative technology. Or two companies come together because there are opportunities for them in terms of the technology features that they can offer and put together. And there's been a little bit more along the lines of, [for example,] Puppet being acquired by a private equity firm. I think it's enabled things to happen like Broadcom-VMware, that wouldn't necessarily have been possible without some changes to market valuations. But that that sort of consolidation within industry spaces like security just has not been as frequent this year, as it has been the last two years really, and especially last year.

Black: Beth, what are your expectations -- and dare I say even predictions -- for the rest of this year?

Pariseau:  Honestly, I think security is going to still be top-of-mind. I am not part of our security news team, but I sort of feel like I'm becoming part of it. Because so much of my beat, that's the hot topic right now. Whether it's open source supply chain, software supply chain, open source security in general and things like open software security foundation that are trying to find systemic solutions to some major risks in open source and all the dependencies everyone shares now.

You could never totally prevent or eliminate risks, but, as an industry, tech really has to figure out cybersecurity. In part because there's now a legal mandate to do so with the President's executive order. For many organizations, they have to figure out zero-trust architecture for the cloud era and supply chain issues. I think you will be hearing more about big data breaches and outages because of security and there hopefully will be more community initiatives around it, and ways that people are shoring it up that they can share with others. To me, that is still going to be a hot topic for the foreseeable future.

I think also developer experience is become a bigger topic. Now that people have made the transition to DevOps and cyber-reliability engineering, in practice and organizationally there's a little bit more of a pattern where platform engineers and site reliability engineers see themselves as product owners within an organizations of a DevOps platform and see developers as their internal customers, and they're starting to apply some of the same customer experience/user experience principles to serving those customers as they do external ones with their company. So, that's bringing about a lot of interesting changes in terms of how people. in the markets I cover work day-to-day, how products are delivered, and the expectations that people have for how their organizations are going to function.

And I just think cloud native technology innovation is still continuing. I think there are things like serverless computing, function as a service, WebAssembly is something that is starting to filter in around the edges as edge computing gets bigger. Service mesh is something that is starting to really come into its own. And, I'm sure, things that I have not even heard of yet by the time we get to KubeCon later this year.