¶ AI Security and Cloud Connectivity Solutions
When we say AI security though securing the LLM like firewalls or like , what do they mean by security ?
Yeah , that's a good point . It doesn't mean the data . It more means like guiding the development of AI to not kill us as humanity , that kind of AI safety , more like a safety .
Yeah , yeah , yeah , okay .
Like human safety yeah .
Welcome to the Cables to Clouds podcast , your one-stop shop for all things hybrid and multi-cloud networking .
Now here are your hosts , tim , chris and Alex .
Hello and welcome back to another episode of the Cables to Clouds fortnightly news . Hello and welcome back to another episode of the Cables to Clouds Fortnightly News , the news where we take a bunch of news articles and put them in one big old battle royale . So this week we've got some good ones .
Of course , joining me always are my two co-hosts , chris Miles at BGP Main on Twitter and Alex Perkins at Bumps on the Wire on some other platform , something else , something else , some bullshit . So we'll just launch right into the news .
Of course , as we always pointed out when we cover these news articles , obviously myself and Chris work for Aviatrix , a cloud networking provider . So the reason I bring that up is because we have a couple new articles from Aviatrix . We shouldn't be afraid to cover them . But of course we also want to make it obvious that we do work for those places .
So the first is a new offering , new solutions offering from Aviatrix , which is called the AWS Cloud WAN Connector , and it was really cool internally to see this kind of be developed . This was developed by our professional services organization and what's really cool about it is there's a lot of ways to connect to Cloud WAN .
I mean , we've talked about Cloud WAN a hundred times on the show here . Obviously we had Ruskin Dantra from AWS on to show us CloudWin . We did you know a cloud demystified series all about CloudWin , so we followed it pretty much since the beginning .
But what's cool about the CloudWin connector is that it is an orchestrated way to set up a way for CloudWin to connect to a third party , in this case through Aviatrix .
Cloudwin has always been extensible to pretty much connect to any kind of a third party using the same kind of connectivity options as TGW Connect or using TGW attachment type of stuff using the CloudWin CNEs .
So what's cool about this solution is that we have a vendor that is automating that process of spinning up not only their connectivity but also orchestrating the CloudWin side of it , which I don't think we've really seen so far .
So just the idea of vendors leaning in and stepping above and starting to orchestrate both their stuff and the cloud native stuff , like the more complicated stuff like CloudWin , I think is a really good kind of indicator of where things are going . You know , businesses really want easy in the cloud and it's been hard to get there , especially with integration .
So I think this is a step in the right direction . What do you guys think ?
This seems like a really interesting solution and I don't think there's really many other people that are doing any kind of automation integration with CloudWan , right , I guess . My question is so does this automatically if you want to extend this right ? Obviously they're using Aviatrix so that you can extend into other clouds . Yeah , how does it do ?
It's like if it spins up a new segmentation within your platform , does that just get automatically extended ? It's like if you have a footprint in Azure , like how does it work to ? Can you basically take those regions and segments within CloudWin and kind of extend them out through your platform ? Is that like part of the general idea ?
Yeah , so that was kind of the area I was going to go in , so thanks for calling that out .
Alex , yeah , you're welcome .
Yeah so the one thing CloudWin does very well is you're able to get this VRF-like behavior in the cloud . You have these network segments that can be multi-region et cetera , right , so you can build out these segments to keep network segmentation . It's been a thing for 30-plus years at this point , so it's nice to see it finally in the cloud .
I'm surprised it took this long . So it's nice to see it finally in the cloud . I'm surprised it took this long . But the CloudWare and Connector , I mean obviously that's great for environments within AWS , right , but extending that segmentation outside of . AWS can be extremely difficult . You have to do a lot of manual work .
I'll say it's doable , but it's a ton of manual work and it's a lot of overhead to manage . So one thing that is cool is , since Aviatrix now has this integration , like you said , new segments that get stood up automatically have a multi-cloud context . Right , we can .
We can ingest that data about the segment , map it to a network domain , which is essentially what we define as our network segments , and that extends in a multi-cloud context . Right , so you can have that same segment existing from a connectivity perspective within Azure or GCP . I keep saying GCP . I don't even think they're going by GCP anymore .
I think it's just Google Cloud , which is hard .
Yeah , we need it , we need it .
We need a short one . The acronym was great , so you guys are dropping the ball on that . I think . Google Cloud , OCI , etc . A really cool integration and I used to work in our PS organization who helped develop this and I know a lot of the guys that worked on this and it's cool to see this finally coming to fruition . It's very cool .
Yeah , I mean , what's really cool in the , in the article also , is that mentions actually I don't remember if it mentions in the article or not , now that I haven't read it in a couple of days but it's built on a , on a framework , not just , not just like we didn't build it one-off for CloudWan , right , we're building a whole extensibility framework for every
possible type of way to integrate , right ? So one thing that's lacking in the cloud is standards . As we all know , right , like , every CSP has their own way of doing things .
So , kind of our philosophy is like , let's be extensible and let's consume all the clouds in the way that they will , let us do it and let's extend that , so not to harp on it too much . But I really think that's the philosophy that's going to really make multi-cloud work . So , okay , one more from Aviatrix .
Again , it was just because it just happened to work out that Aviatrix like released multiple things this week that were within the last week , that were kind of worth covering , that were stuff we would , we would talk about , uh , you know , even if they didn't come from Aviatrix .
So this one's an interesting article and , uh , it's press release from Aviatrix and it talks about how , uh , aviatrix just kind of did a uh , I hesitate to say case study I don't think the full case study is actually done yet , but uh , but has done a lot of collaborating with a really big customer on their Aviatrix deployment and kind of how they're consuming
Aviatrix , and did a like-right , apples-to-apples kind of comparison and decided that , you know , hey , if we did , if we get the same functionality using just kind of native cloud constructs , given , I think , largely in data transfer costs alone , you know there's a huge uplift .
So the press release says that basically , after doing the math on this , that they've been able to save this customer 25 million , you know , just in cloud costs . Now I look at these things with a huge grain of salt from anybody that releases numbers like that , cause that's a , that's a sticker shock number to me . So I actually was like what ?
So I actually went inside and I looked and I actually was able to dive in and get in to the spreadsheets , look at the data , see how we came at the numbers and , um , I mean I don't know what to say , like it was legit .
I will say , without again talking about the specific customer and getting their specific deployment and getting into , obviously , names and NDA liable stuff . You know the way they are deployed and the traffic patterns they use and the services they're consuming obviously contribute a huge amount to that number .
Right , I would not go and say every single customer that Aviatrix has is going to save $25 million in the cloud , right , and I think I would say that of Valkyra , of Prasmo , of any provider of it , of any provider out there .
But having said that , right , if you have really high data transfer , heavy workloads and have specific kind of traffic patterns , yeah , I mean , those kind of numbers do exist . So it was all legit . It was really cool to be able to go validate that instead of just having to give the side eye to a press release like that .
So what do you guys think about that ?
just in general , yeah , I think it's funny because before this was even in our thing to cover , I think I even asked you guys specifically something like is this for real , Can you guys really do ? It costs you guys $5 million to do . What takes this company $30 million , right , which is where that $25 million save comes from . Yeah , that's right .
And then , reading through it , there's even a mention of I think it said something like due to NAT gateway charges , there's a potential for another $5 million to be saved as well . It's just crazy numbers , though . Saving 25 million is ridiculous . That's just crazy .
Yeah , I think and , tim , you kind of made this point already but I think it all kind of ties back to the fact that , within the cloud , consumption models matter a lot , and this is I've said this a million times but , like , the way that you think about networking in the cloud and on-prem is completely different .
Just because they've added this , this functionality of data transfer , that that you pay per gigabyte for , and that's not , that's not a knock at any particular CSP at all , because they all do it and I get why they need to do it . But , as operators , a lot of times that is obfuscated .
It's very hard to predict and you know at the end of the day , you know , I would like to know how some of these conversations actually happen at the . You know , at the end of the year , once , once companies have really evaluated what they spent on data transfer in the CSPs , you know how they forecast that in the future .
But , to Tim's point , not every company south egress , et cetera A lot of this stuff really matters where it didn't used to matter . So it's just yeah , things like this are obfuscated from you for a reason , I think right .
Yeah , hidden in a line item and all that Exactly . Cloud networking doesn't have a line item on , I think . Right , yeah , if you've looked at Hidden in a line item and all that Exactly .
Cloud networking doesn't have a line item on a bill , right , it's spread across , you know 10 , 20 different things right , a bunch of services .
Sometimes it's , you know , even when we had DeJuan on , he talked about like knowing how to read your cloud bill is very important , because he was talking about things like NAT gateway charges , that falls under within AWS , that falls under EC2 , not , you know , within , you know transit gateway or anything like that .
So it's like it's kind of hard to get everything in one bucket .
Not even just Duan with AWS , but Woody brought this up with Azure as well recently too , just last week , right , I mean , it's definitely something you have to watch out for , and architecture matters a lot .
That's the architecture really matters , and I think people are starting to figure out that architecture matters . But , like , there's still a question of , well , how do we fix it ? Like , how do we build the architecture to make it most effective , especially on the network side of the house , right , there's been a lack of scrutiny on that .
It's just been like , okay , well , we've got to pay this bill because we're doing this business .
And it's funny because , just generally speaking , I actually see a lot of people , or a lot of customers and things like that , that were early cloud adopters . Most of the time it's in AWS . Australia is very heavy in AWS . It seemed like customers and things like that that were early cloud adopters , especially this is most of the time it's in AWS .
Australia is very heavy in AWS . It seemed like a lot of early adopters had these very inefficient models for connectivity , a lot of them not even using things like AWS TGW .
So you'll see , like numerous you know VPCs deployed that all have private VIFs over a single Direct Connect or something like that , and then they start hitting limits on number of VIFs that they can have and then they're wanting to come move to things like TGW .
But it's hard to weigh what that traffic cost is going to look like , because then you've got you know your your Direct Connect egress charges . You've got your TGW data transfer charges , you know if you add a security VPC in there , you've got your GWLB charges and endpoint charges and it's like it's a lot , it's it adds up quick .
¶ AI Security and Amazon AI Development
But yeah , I think we've , we've kind of beat it to death . But yeah , cost cost we've , we've , we've kind of beat it to death . But yeah , cost , cost matters . It's never sexy , it's not the funnest thing to talk about from someone that's , you know , really interested in technology .
But it's a sad reality that we really have to consider it day in , day out here in the cloud .
Yeah , for sure . Okay , I'm going to hand it over to Alex .
I think you got the next couple of uh yep . As usual , we have a storm of uh ai content that we , that we have to go through um , so you'd think by now that I would have learned how to pronounce these names , since they keep coming up . I have not . So , uh , ilia setskeeper and jan leakey we've talked about this before . They recently left open ai um .
Jan went to Anthropic and he heads like a security team at Anthropic , which was a direct competitor , but now Ilya has decided to make his own company called SSI , and that stands for Safe Superintelligence Incorporated . What a name that is . There's not a whole lot of meat to this . This article is from TechCrunch .
It basically just talks about the fact that Ilya left and is now super focused on creating this safety company for AI . What I don't get is how do you make money off of this ? This isn't like providing an AI as a service . This is just making AI secure . So is this going to be a nonprofit ?
It doesn't have any of these details in here and I'm real confused about how you make a company that just does AI security but isn't aligned to any particular company .
Do you guys have ?
any thoughts here ?
When you say AI security , though , you mean just securing the LLM , like like firewalls or like . What do they know ? What do they get ? Not the data , yeah Right .
That's a good . That's a good point . Not , it doesn't mean the data . It more means like guiding the development of AI to not oh , ok , as humanity Right .
That kind of AI safety yeah , more like a safety . Yeah , yeah , ok .
Like human safety ?
Yeah , exactly . No , I don't know . I mean , like somebody needs to do it I . The government obviously is going to be completely worthless at it because our government is full of people who are too old to , you know , do anything about anything , Not even get into the debate thing . But that's the big one Like .
So if we can't trust the government , who can we trust ? And I'm not sure for profit , nothing for profit could be the people we trust either . So I don't know , Maybe that's the answer . I don't know .
I mean , I think you called it out right there . I think the key word in this entire thing is safety .
It's kind of what they're leaning in on , because they I think I can't remember which one of them said it , but one of them was talking about like you know , they're , they're expecting AI that is superior to humans uh , to be uh , you know , available with I think it was said within the decade , um , and it's really just about being prepared for that .
Um , yeah , so it's .
I mean , and that was in 2023 . Yeah , that was a blog post that Ilya published with Jan in 2023 .
And I wonder how much of this is just like . Obviously there's a bit of a jab at Altman here calling it , say , super intelligence , right , whereas that was kind of the whole crux of the kerfuffle that happened whenever he left OpenAI . But yeah , I don't know . I mean I'm glad the thing is .
One really good takeaway from this is , I think this is a company that I think they even stated in the article will have no issues getting funding . This will be something that will definitely get plenty of money to develop exactly what they want to do Just the integration , just the , the integration and the .
I guess I don't know if it's going to go into helping build regulations and shit like that , but all I can say is I hope it's successful , cause I think we're all in agreement that you know safety and you know the emergence of Skynet has us all a little concerned , so we'll we'll have to see .
But it should be as like an advisory council to an actual legislative body . Exactly that's what you need , right , yeah ?
that's true . Yeah , that's what I was hoping that this kind of was . I was hoping that's the direction they were taking . It was a little alarming . I was trying to find it in the article , but the guy who mentioned that they won't have any problem raising capital his last name's gross .
I cannot remember who his first name was , but I looked up this person and they were part of a board of another company . I'm so confused at what they're raising money for . How are they going to make a profit ? You know what I mean ? I just don't .
Yeah , who are the ? What are the investors investing in exactly ? Are they just ?
covering what is covering the spread , basically , yeah .
You know ? Yeah , I don't know .
All right , we'll move on from that . So the next one . Finally , aws is really going to throw their hat into the ring here . This is an article from read writecom and it's a report that Amazon is developing a killer you know in air quotes killer AI chatbot that will directly compete with chat GPT .
Looks like it's been codenamed Metis and it's powered by an Amazon LLM known as Olympus , which they say is being trained on a two trillion parameter two trillion parameters of data , which is around twice the capability of GPT-4 . That makes me wonder , like I wonder what these parameters are that are being chosen to be trained .
Yeah , just , it's a ridiculous amount , but it doesn't . It's again , it's like a couple paragraphs . It's not super detailed , but this is their goal , right Is to take on their competitors and kind of try to catch up in the AI space . And then , funny , the second half of this article is titled Jeff Bezos questions Amazon's position .
It's about he sent an email to senior staff asking why Amazon isn't being used for AI workloads . I just thought that was kind of funny .
What are you talking about ? They're the leader in AI . That's what I hear all the time anyway . Yeah , dude , everybody uses Q . Oh yeah , yeah , dude , actually this is just going to go up and up and up and , as it is , we're already having power grid problems from the amount of resources . Think of like you thought Bitcoin was bad ?
Now this is absolute insanity , dude . How long until we're doing rolling blackouts so that somebody can generate some dumb picture , some dumb AI-generated picture or something ?
That's something real quick before you jump in , chris , that's something we need to talk about at some point , as I keep seeing nobody's really talking about the demand of power that is having to come because of all these ai .
You know work and cooling everywhere , like there's so much , yeah , so much , so many resources being thrown at this shit but there's been like regulations for nuclear power now that are suddenly like moving , so I don't it would be an interesting topic to talk about at some point . Go ahead , chris .
Yeah , um , man , that would be . I'll tell you right now . Uh , there's , I called it out earlier . There's one thing I don't really enjoy talking about . That much is cost power and cooling is close behind that . Uh so maybe I'm already sick of talking about chips and , uh , ai becoming quite a long list , but maybe we'll have that discussion .
But yeah , I echo everything you guys just said
¶ Telco Networks and Hyperscalers
about this article . One thing that I think that is quite funny I don't know if this is a Freudian slip or whatever , but one of the sentences in here says Metis will enable Microsoft to take its place among other big tech players such as Google and Microsoft .
And it'd be so funny if they were just like yeah , metis will be so bad , it'll make Microsoft even more money .
If it wasn't a quote , it's probably a mistake on the part of the person that wrote the article , but it also might betray their . It might be a true Floydian slip for what their opinion is as well .
Also , real quick Amazon , as they call it , might be a true Floydian slip for what their opinion is as well . I don't know . Also , real quick Amazon , as they call out , is also backed Anthropic .
So it's kind of weird .
It's like they're coming out with their own thing , but they also backed a different competitor as well .
That's not super unusual . Cisco does that crap all the time , right . How many times have they bought a company where they were creating a product ? I'm sure the plan is that , hey , we'll catch up faster by taking Anthropic stuff and using it with us , or whatever .
It's always a question of are they going to gut it and try to make it work with what they've got , or are they going to be like , hey , they're actually further than we are , we'll drop this and we'll do that ? I guess who knows ? All right .
And lastly , it says there's a proposed launch for this Metis chatbot in late September . I would call BS on that . I bet if it's going to get announced it would be at reInvent .
Yeah , well , maybe they might be thinking like we can't wait that long .
Maybe mean that's end of november , that's true , that's yeah , one thing I did , so this is this is an emphasis on amazon putting this out right yeah , so then , but if it's directly marketed at consumer um what chat gpt is . Uh , that's a good point might not matter at all . Yeah , um , so yeah , well then it could be september well , no , who knows .
Well , we'll cover it either way , I'm sure yeah , yeah , we should , uh , we should ask , ask it to design a network for us like yeah , chat , gbt we should revisit that .
Yeah , we should probably revisit that I'm sure once every every 12 months . We'll do it and we'll see if it gets better um and we'll start getting to ask maybe then we'll just get it to write the fucking show for us . Fuck that . Yeah , there you go . Just kidding , anyways , okay .
So moving on , I have one here from Fierce Network talking , all given by Colt Technology Services , which is based in the UK , pushing this idea of tech as hyperscalers . So , basically , the overall sentiment is that you know telco networks take a long time to provision and a lot of that is based around the fact that telcos need to over provision , right ?
You know so if you think about your common data center , right , you're usually what . I've already been out of the game so long now . I already forgot what it's typically like . It's like three to one or one three , whatever . Typically your oversubscription model within a data center being that you know you have bandwidth allowing for peak times when you need it .
Telcos have to , kind of it's exacerbated for them , right ? So they need to account for that and they recognize that that's something that the cloud providers or these hyperscalers have perfected the idea of elasticity and agility in that particular facet . So they're talking about .
You know telcos can maybe shorten their time to market with provisioning services by leveraging things like cloud services to run over the top . You know it kind of gets into this whole idea that it's easier said than done . And you know it kind of gets into this whole idea that it's easier said than done .
And you know , like I think all of us talking from a technical perspective can Like , as soon as I hear that , like alarms are going off in my head like how the fuck would you like , how would that be cost effective , how would how would that be enough revenue to account for what you're going to pay to the underlying CSP ?
I mean , unless there's you know there's always , you know , deals happening behind the curtain , right , so maybe something could be , could be negotiated in that fast .
But like we just talked about data transfer charges for you know 10 minutes and like that at a telco level scale is going to be insane and I don't know exactly how that would be cost effective for a telco . So , yeah , I mean I could probably keep talking about this , but I'll defer to you guys and give your thoughts .
Maybe the telcos will finally realize some of that economy of scale that we were all promised .
Maybe that's on their slides when they presented this . Eventually , it's going to be cheaper .
I've always had this idea that telcos and CSPs could kind of work together , just because telcos already have like all these edge presence everywhere . And it's just been surprising me that this hasn't happened .
Like it allows the hyperscalers to push out and then it would allow the telcos to like use this global spanning network and it would just be like a mutual deal and they could save so much , like both sides could save money on this . I've just I've always been surprised that this hasn't happened and instead we see things like what's the AWS wavelength right ?
Like they're starting to try to push out their own stuff , but they have outposts Like why couldn't you put like a region , like small region ? What are they Local zones , right ? Is what they're called with AWS ?
Yeah .
Why ? Why couldn't you have one on a cell tower and then , like , use the network to expand out ? It's just something that I've always wondered why they don't do . And I know Azure has gosh , they call it like Azure for operators , I think is what it's called , and that's kind of how they interact with the telcos . Just surprising to me .
And as far as I know , dish is the only one that has tried to move their core network into a hyperscaler . And when's the last time you heard about DISH ? Because this was like years ago and they had moved their core network onto AWS . Has anyone heard about anything about this in years ? Probably not right , so I don't know .
Yeah , I'm kind of surprised that the hyperscalers and the telcos didn't partner more . I mean , I guess , like you said , you've got Wavelength . What is it Kuiper Like from ? You know , kuiper is another way to connect .
You know , you got AWS just pushing out all these different ways to connect so that they can deliver what they want to deliver , which is , of course , the cloud services on top of it . So I would have thought partnering with telcos , but I actually .
It actually doesn't surprise me that much and and the reason is because of what's in the article , right like , think about being abs and going to the telco and being like I'd love to use your , your backbone for this , and they'll be like , okay , well , you know , that'll be three 90 days wait period for for this to happen . No agility right .
They might have even gone to them first way back in the beginning and said , hey , what would it take to do this and realize the timelines were just not conducive to their business model .
Yeah , I mean you see it a bit , but it's always centered around private connectivity . If you think about the MPLS providers , that's a common one , but it's always very customer-focused , one-for-one relation . It's never in the core backbone of the telco networks and you see it a bit from the vendor side , probably more often than the telcos .
I think Cisco had the deal with Google for the NCC connectivity thing . Yeah , I'm surprised this didn't happen sooner . I agree with you , alex .
Yeah , and also just to speak to some of the challenges that they have . Like there's a reason that OpenStack is still being used very heavily in telco networks and there's things like OpenRAN the functionality that telcos need in their networks the functionality that telcos need in their networks .
It's been probably 10 , 15 years that people have been trying to virtualize telco functions or even containerize them more recently . So that's got its own whole set of challenges that they would have to figure out how to run on these hyperscalers as well .
Yeah , agreed . Lastly , we'll round this one out with our favorite person . To talk about the favorite , I'm talking about 37 Signals , which is obviously the company he works for , which is responsible for software such as Basecamp and Hay . To be honest , I've never used either one . I know they're relatively popular . I'm just talking about the name of the article .
Is Our Rainy Days Ahead for Cloud Computing ? How long was just like dude this has been ? How long has it been saying this shit ?
He's got one talk track and he's going to roll with it .
I mean no disrespect to the guy . I mean obviously he's probably regarded as a god over at that company . But I think he's been talking about repatriation for years at this point right , and how the cloud was not cost effective , not conducive to their business . And now this most recent article is saying I think you said , alex , this year they're on target .
To what is it ? I wouldn't say revenue , but you're making an additional million dollars in profit . And I'm like man is all this in the media and we're talking about $1 million in profit .
Right , which it's not insignificant .
No , it's not . I'm not going to say that's not , but like fuck man , I thought it was going to be in the tens of millions .
Right ? Well , what I've always taken taken issue with is his argument that the cloud doesn't work for just about anybody . He doesn't say hey , we looked at the way we do business and the way we do business doesn't work with the cloud . Because you know what ?
There are a good number of companies out there that the work , the thing that they do , the apps that they have are static and don't have elasticity , which we've said a hundred times . That elasticity is if you're never going to save money in the cloud , that's where you're going to save it .
So there's only the elastic , uh , elasticity being able to draw down when you don't need it , build up when you need it and only when you need it . Right , that's the , that's the , the , the dream , um , but like he doesn't just say like , hey , basecamp , it never made sense for us . He just always says Basecamp and basically everyone else can't .
He just speaks for everybody . I think that's always fine to be the most frustrating thing . Most companies out there can't benefit from the cloud . I didn't realize he was on the board of most companies , that he had all this insight into how they do their business . Right , that's what I always took issue with .
Yeah , no , I mean , I fully agree with that and it's funny because I'm reading it right now . But in this article it says was using the cloud faster ? And the answer is yes , but it didn't matter , and that they still use the cloud for experimenting with new products . So it's like yeah , that's the whole point ? Exactly , that is .
The whole point is that when you start in the cloud and then you're crazy . You're crazy if you don't start in the cloud and you're crazy if you stay in it , and it's just . That's never the point of when he has these articles .
It's just look at how much we saved and you can too , without going into any of the nuance behind it and how that there's not a one-size-fits-all answer for everybody yeah , totally all right .
Uh , I'll go ahead and wrap us up now . We do have more articles and , uh , they will be in the fortnightly news , which we , you know , have in every uh , episode of fortnightly news . We can't cover all of them , but please do take a look at some of the other articles that we didn't cover .
We do think everything we put in there is relevant and interesting , and with that we'll go ahead and wrap it up for this week .
One quick note I'll say is we just so ? I know many of our listeners out there are listening on Spotify and we've been lazily not looking at it , but I just set up a Spotify for podcasters account yesterday and I've noticed that you know they offer this option for you know commentary from listeners about what did you think of this episode ?
So if oh , that's awesome .
Yeah , so if you want to fill that out and just tell us a little bit of feedback on how you felt about the show , what you know , what you , or just general commentary , we'd love to hear from you . So if that's an easy mechanism for you to use , talk to us , baby , let's have a chat . I'd love to hear from you .
That's awesome . I didn't know that existed . That's really cool . I wonder if any of the other ones out there we should do more work on that , but yeah , that's really cool . Thanks for bringing that up . Speaking of , if you did like or hate what you heard or saw today , please follow us on all the pod catchers , obviously , and also .
I love the idea of gaining a follower out of hate .
Oh , I mean spite , spite . Following is like that's , that's really exciting .
That's the new thing these days , yeah .
Hate engagement is like 10 times as powerful as like engagement .
Yeah , people have been doing it on Twitter for years . Like you see people just talking shit to people and then they're , they're following the , the celebrity that they're shitting on .
you know it's , yeah , it's like the the Howard Stern movie or whatever , where it's like you know the average . Average person that hates him listens longer than the average person that likes him . They want to see what do you say , so anyway , yeah , so thanks for joining us and we will see you next week .
For our YouTube channel to be notified of all our new episodes , follow us on socials at Cables2Clouds . You can also visit our website for all the show notes at Cables2Cloudscom . Thanks again for listening and see you next time .