Changing jobs....

[Scarpozzi]

You've probably seen a few posts from me. There is just not enough SharePoint talent going around.

I always wanted to work with Sharepoint implementation, but I was the Linux admin so I had my hands full keeping the ERP and systems monitoring software running. I worked for 2 large enterprise organizations with around 20k user accounts at any given time. I had the windows admins decide against Sharepoint due to the implementation struggles. My assumption is they didn't want to write the code required to backfill missing data in Active Directory from Human Resources...or maintain it moving forward. (required groups and attributes, etc) I'm pretty well versed in bash scripting and dabbled in in vbscript...but Powershell is much easier.

I'm still wondering what makes Sharepoint so difficult to manage that 3 admins I know gave up on it before ever getting it off the ground.

 

[IndyColtsFan]

I always wanted to work with Sharepoint implementation, but I was the Linux admin so I had my hands full keeping the ERP and systems monitoring software running. I worked for 2 large enterprise organizations with around 20k user accounts at any given time. I had the windows admins decide against Sharepoint due to the implementation struggles. My assumption is they didn't want to write the code required to backfill missing data in Active Directory from Human Resources...or maintain it moving forward. (required groups and attributes, etc) I'm pretty well versed in bash scripting and dabbled in in vbscript...but Powershell is much easier.

I'm still wondering what makes Sharepoint so difficult to manage that 3 admins I know gave up on it before ever getting it off the ground.

In terms of on-premise implementations, I am not sure I’ve ever seen a properly implemented and governed installation unless I was the one who set it up. It’s a complex product to install, understand, and maintain. And to build a properly scaled environment is something few can really do. For example, I was doing some independent consulting and got a call from the VP of a local consultancy asking if I‘d take a look at a local company’s install. This company had 6-8 Sharepoint 2010 web front ends and they let some complete idiot “consultant” convince them one would be enough with SP 2013. The problem? The farm was publicly accessible and had around 600,000 user accounts and with up to 30,000 concurrent users. The “consultant” was obviously an idiot but the IT management was even dumber for not booting his ass out the front door at such an incredibly obvious bad decision and advice. Yes, I fixed it and yes, they loved me for “saving” their jobs. They should’ve all been fired anyway (and several eventually were).

On top of that, companies (especially small ones) tended to develop all sorts of custom shit and make it the system that runs most of their critical business processes. The issue with this approach is that it paints you into a corner which is very expensive and hard to get out of. I had a client like that late last year and earlier this year and it got to the point where I made the call to rip the bandaid off and fix issues after the migration. They had NO resources who understood SharePoint and that was inexcusable given the processes it ran for the company.

For my part, I’ll be honest - any company still leveraging an on-prem version of SharePoint who isn’t planning on moving to SharePoint Online is short-sighted to the point of stupidity. I am pushing my management hard to abandon any offering which results in an on-prem end state. We don’t (fortunately) do many of them, they’re not worth the hassle since most are tiny clients who have awful environments, and it isn’t something worth us devoting time to keep our skills current. And now Microsoft is bringing out yet another on-prem version which will (wink, wink) “be the final one.” They should just stop already.

 

[Scarpozzi]

In terms of on-premise implementations, I am not sure I’ve ever seen a properly implemented and governed installation unless I was the one who set it up. It’s a complex product to install, understand, and maintain. And to build a properly scaled environment is something few can really do. For example, I was doing some independent consulting and got a call from the VP of a local consultancy asking if I‘d take a look at a local company’s install. This company had 6-8 Sharepoint 2010 web front ends and they let some complete idiot “consultant” convince them one would be enough with SP 2013. The problem? The farm was publicly accessible and had around 600,000 user accounts and with up to 30,000 concurrent users. The “consultant” was obviously an idiot but the IT management was even dumber for not booting his ass out the front door at such an incredibly obvious bad decision and advice. Yes, I fixed it and yes, they loved me for “saving” their jobs. They should’ve all been fired anyway (and several eventually were).

On top of that, companies (especially small ones) tended to develop all sorts of custom shit and make it the system that runs most of their critical business processes. The issue with this approach is that it paints you into a corner which is very expensive and hard to get out of. I had a client like that late last year and earlier this year and it got to the point where I made the call to rip the bandaid off and fix issues after the migration. They had NO resources who understood SharePoint and that was inexcusable given the processes it ran for the company.

For my part, I’ll be honest - any company still leveraging an on-prem version of SharePoint who isn’t planning on moving to SharePoint Online is short-sighted to the point of stupidity. I am pushing my management hard to abandon any offering which results in an on-prem end state. We don’t (fortunately) do many of them, they’re not worth the hassle since most are tiny clients who have awful environments, and it isn’t something worth us devoting time to keep our skills current. And now Microsoft is bringing out yet another on-prem version which will (wink, wink) “be the final one.” They should just stop already.

I came from the world where cloud was a bad idea because all the employees were local and it made no network sense to send that much traffic to a data center hundreds of miles away.

That's all changed now, of course. Quite a few services are slower and people are just dealing with it. Adapting to SaaS is tough for old school admins like me, but I get the business sense of it. Especially for things like Email and other critical systems that require lots of SAN space. I've done enough RAID array repairs, restores from backups, etc...to know I never want to do it again. My org moved to Office365 about 5 years ago and the current CIO has no clue about real IT. She just wants to manage contracts and blame vendors if the technology doesn't work. I guess that's what it's come to just about everywhere.

I'm sure the latest "on prem", as you call it, has something to do with SharePoint being included in some licensing bundle they sold to the government that outlives the versioning product roadmap...so it can't just die.

 

[Spacehead]

the stress is the worst part for me

bad days at work can cause my blood pressure to go through the roof, and my doctor has been harping on me about that already

I'm pretty easy going by nature but, yeah, work can bring out the stress at times. Sadly, anymore i have to keep reminding myself i've taken an " i don't give a fuck" attitude. If i don't give a fuck i don't get mad.
I'll do my job as best i can but i'm not going to get upset when the boss harps on me to "hurry up & get that job done!" that just landed on my bench & was due in June. Not my problem it's that late, it's yours.

 

[IndyColtsFan]

I came from the world where cloud was a bad idea because all the employees were local and it made no network sense to send that much traffic to a data center hundreds of miles away.

That's all changed now, of course. Quite a few services are slower and people are just dealing with it. Adapting to SaaS is tough for old school admins like me, but I get the business sense of it. Especially for things like Email and other critical systems that require lots of SAN space. I've done enough RAID array repairs, restores from backups, etc...to know I never want to do it again. My org moved to Office365 about 5 years ago and the current CIO has no clue about real IT. She just wants to manage contracts and blame vendors if the technology doesn't work. I guess that's what it's come to just about everywhere.

I'm sure the latest "on prem", as you call it, has something to do with SharePoint being included in some licensing bundle they sold to the government that outlives the versioning product roadmap...so it can't just die.

I started in infrastructure and had a hardcore hatred for all things cloud. I used to even advocate for the on-premise version of SharePoint because it was (and in some ways, still is) more powerful. I still have a large server here at home with 50-60 VMs and several SharePoint farms.

With all of that being said, over the years, my view changed. What I said in my previous post wasn’t said out of arrogance, but out of observation of hundreds of SP installs - none that I have seen were built 100% correctly except ones I built myself. Obviously some were pretty close and pretty good, but most were really bad. And even in relatively well-built environments, maintenance is a serious pain. You really have to have someone with SharePoint and SQL knowledge along with someone who has a good understanding of Windows and exceptional troubleshooting skills to resolve issues. Patching SP environments, while not as harrowing as in the past is still not fun. So it boils down to needing 2-3 very skilled (and likely highly paid) people to run SharePoint and unless you’re a very large business, that’s hard to swallow. So from a cost perspective, feature perspective, and a maintenance perspective, moving to SharePoint Online when most businesses are already paying for it makes perfect sense.

Now, I have some degree of skepticism about the costs of the cloud in the longer term. I can see vendors getting everyone locked in and prices climbing to the point where an on-prem infrastructure might be cheaper. At that point though, it would probably cost a company millions to move everything back so they’re pretty much stuck. Moving between clouds would also be very expensive, so Microsoft has them exactly where they want them. And even in the next on-prem version, Microsoft is throwing down the gauntlet - it is called SharePoint Subscription Edition for a reason and MS is determined to get that recurring revenue come hell or high water. I look for the same to happen with any on-prem versions of other server products as well.

 

[cytg111]

I came from the world where cloud was a bad idea because all the employees were local and it made no network sense to send that much traffic to a data center hundreds of miles away.

That's all changed now, of course. Quite a few services are slower and people are just dealing with it. Adapting to SaaS is tough for old school admins like me, but I get the business sense of it. Especially for things like Email and other critical systems that require lots of SAN space. I've done enough RAID array repairs, restores from backups, etc...to know I never want to do it again. My org moved to Office365 about 5 years ago and the current CIO has no clue about real IT. She just wants to manage contracts and blame vendors if the technology doesn't work. I guess that's what it's come to just about everywhere.

I'm sure the latest "on prem", as you call it, has something to do with SharePoint being included in some licensing bundle they sold to the government that outlives the versioning product roadmap...so it can't just die.

Ill never surrender to the cloud. Eff that. Single Point Of Failure for Everything. SPOFE.

 

[Scarpozzi]

Ill never surrender to the cloud. Eff that. Single Point Of Failure for Everything. SPOFE.

That and not knowing what your critical stuff is actually running on. For me, I can't stand the thought of putting a database on a virtual machine that's sharing resources with an undisclosed number of systems. I also worry about hypervisor network security. My organization recently moved stuff to the cloud and we have all these intermittent security issues when accounts are updated. I think it's likely a timesync issue between middle tier servers or something, but no one can visualize what they're working with because we had so many people leave last year. (I don't work in IT-proper any longer, so not my problem)

It's all about costs though....human resources is one thing. Maintaining large-scale infrastructure like backup batteries, generators, SANs, and tape libraries is another.

Most companies have servers in non-ventilated closets and want to buy hosted solutions these days....just so they don't have those sunk infrastructure equipmen

t and service costs that come up constantly.

 

[Red Squirrel]

Yeah I don't like the idea of cloud, even for my own personal stuff. I want to have full control and knowledge of where my data is. I can see the NAS, see the individual drives, I can login to it and look at the raid status, I can see the individual files, and I can also see the backups, and see that the backups are working. Not a fan of the "it just works" philosophy where you don't see what's going on and just hope it never goes down.

Same thing with programs that save stuff in a weird format that is not just a file. Like for example, local finance or photo software that uses it's own built in database. If I can't see individual files and KNOW that it is my data, but it's just somewhere "in the computer" then not a fan of that either. How do I back that up and how do I ensure I can get that back if I clean install.

 

[RearAdmiral]

That and not knowing what your critical stuff is actually running on. For me, I can't stand the thought of putting a database on a virtual machine that's sharing resources with an undisclosed number of systems. I also worry about hypervisor network security. My organization recently moved stuff to the cloud and we have all these intermittent security issues when accounts are updated. I think it's likely a timesync issue between middle tier servers or something, but no one can visualize what they're working with because we had so many people leave last year. (I don't work in IT-proper any longer, so not my problem)

It's all about costs though....human resources is one thing. Maintaining large-scale infrastructure like backup batteries, generators, SANs, and tape libraries is another.

Most companies have servers in non-ventilated closets and want to buy hosted solutions these days....just so they don't have those sunk infrastructure equipmen

t and service costs that come up constantly.

You can always buy dedicated resources at a cloud provider yes?

 

[Scarpozzi]

You can always buy dedicated resources at a cloud provider yes?

Yep....so they claim. =P

 

[IndyColtsFan]

Ill never surrender to the cloud. Eff that. Single Point Of Failure for Everything. SPOFE.

Single point of failure? How do you figure? Do you seriously think your company has more redundancy than Microsoft, Amazon, or Google? MS replicates the environments between datacenters for service redundancy and has an incredible amount of redundant bandwidth.

Your costs to match the redundancy of a Microsoft or Google far exceeds their current licensing costs. You'd have to pay the salaries of highly skilled people to design, build, and maintain it. You'd have to pay for a lot of expensive hardware, software, and bandwidth. You'd have to pay licenses for a lot of additional software as well. The fact is, if technology is not your company's core competency, there is no way you can achieve what Azure, AWS, and Google Cloud can achieve at their price points.

 

[Scarpozzi]

Single point of failure? How do you figure? Do you seriously think your company has more redundancy than Microsoft, Amazon, or Google? MS replicates the environments between datacenters for service redundancy and has an incredible amount of redundant bandwidth.

Your costs to match the redundancy of a Microsoft or Google far exceeds their current licensing costs. You'd have to pay the salaries of highly skilled people to design, build, and maintain it. You'd have to pay for a lot of expensive hardware, software, and bandwidth. You'd have to pay licenses for a lot of additional software as well. The fact is, if technology is not your company's core competency, there is no way you can achieve what Azure, AWS, and Google Cloud can achieve at their price points.

Single Point of Failure is your network (or theirs....or any route between you or them). That's what I took that to mean.

If you're in a mission-critical situation where you can't be down, you need redundant dedicated networks and that gets pricey. If you're in a big city, it's likely more viable. My organization had a network they used that was run by AT&T. The route to AWS went a couple of miles out of the way because of where it terminated. It was hopping between Memphis and Atlanta before going to DC and back again....lots of latency potential.

 

[IndyColtsFan]

Single Point of Failure is your network (or theirs....or any route between you or them). That's what I took that to mean.

Yes, that is likely what he meant, and it is a suspect argument for a few different reasons:

1. Redundant internet connections aren't terribly expensive to implement and the ability to eliminate on-premise email, file, and collaboration servers more than makes up for it.
2. Let's say your internet connection drops - how are you going to communicate with vendors, clients, etc? I mean you still have access to your datacenter (assuming it's in the same building) and can communicate internally, but if you have cloud-enabled services, you can literally go to any live internet connection and you're back in business with both your vendors, clients, and other internal staff. If your internet connection goes down frequently enough for this to be a major concern, I think the thing to worry about is finding better connectivity.
3. I'd wager that Microsoft has greater service uptime for email, SharePoint, etc. than most companies do with their on-premise environment.

These are the kinds of things my clients are discussing. I've got clients with as few as 250 seats to clients in the 300,000 - 500,000 seat range and they're all saying the same thing - they want to eliminate as much local infrastructure as possible, free up IT staff to add actual value to the business, and most importantly, implement services which enable users to work from anywhere with ease. Like I said earlier, I'm really trying hard for us to drop the on-premise offerings in our practice because they're just not worth our time. We can maybe take them and outsource them, but our staff has more important stuff to work on to be honest.

If you're in a mission-critical situation where you can't be down, you need redundant dedicated networks and that gets pricey. If you're in a big city, it's likely more viable. My organization had a network they used that was run by AT&T. The route to AWS went a couple of miles out of the way because of where it terminated. It was hopping between Memphis and Atlanta before going to DC and back again....lots of latency potential.

There are going to be mission critical, niche applications that will always need to remain on-premise. Even those are gradually being located in external datacenters (the Cognizants of the world) because of cost advantages.

There was a good discussion over at [H] a few months ago where a technology manager evaluated his options of on-prem vs cloud for his workloads. He did all of the cost computations factoring in hardware costs, software costs, labor, etc. His finding was that the cost to license everyone for M365 was less than only the software licensing he'd have to pay to come close to that level of redundancy for an on-premise engagement. He still had hardware (and regular upgrades for it) and labor costs above and beyond those to consider as well.

Yeah I don't like the idea of cloud, even for my own personal stuff. I want to have full control and knowledge of where my data is. I can see the NAS, see the individual drives, I can login to it and look at the raid status, I can see the individual files, and I can also see the backups, and see that the backups are working. Not a fan of the "it just works" philosophy where you don't see what's going on and just hope it never goes down.

I promise you, you have nowhere near the redundancy or resiliency of a cloud provider. And personal stuff is one thing - try running a business where technology isn't your core competency. If your house burns down tomorrow while you're on vacation, how well is your infrastructure working then? How are you retrieving the data? Even with my own personal on-premise server, I do both local and cloud backups precisely for this reason. My main server is old (running almost 24/7 for the past 9 years) but it is hard for me to justify upgrading it at this stage. I'll probably end up with a low-end server or desktop with a ton of cloud-backed disk space for large files which need to be stored locally.

Same thing with programs that save stuff in a weird format that is not just a file. Like for example, local finance or photo software that uses it's own built in database. If I can't see individual files and KNOW that it is my data, but it's just somewhere "in the computer" then not a fan of that either. How do I back that up and how do I ensure I can get that back if I clean install.

We're not talking about your personal stuff. In the real world, you don't get to make those decisions. You have input but ultimately the business decides what applications to use and you support them whether or not you like it or not.

 

[jpiniero]

For my part, I’ll be honest - any company still leveraging an on-prem version of SharePoint who isn’t planning on moving to SharePoint Online is short-sighted to the point of stupidity. I am pushing my management hard to abandon any offering which results in an on-prem end state. We don’t (fortunately) do many of them, they’re not worth the hassle since most are tiny clients who have awful environments, and it isn’t something worth us devoting time to keep our skills current. And now Microsoft is bringing out yet another on-prem version which will (wink, wink) “be the final one.” They should just stop already.

Been awhile but the Online version is very limited in what you can do with custom apps compared to the on-prem version.

I'd just rather use network file storage for document sharing. I'm sure there are plenty of companies that don't want to give Microsoft all of their documents.

 

[IndyColtsFan]

Been awhile but the Online version is very limited in what you can do with custom apps compared to the on-prem version.

Not really true these days. You can build almost anything with SPFX and the Power Platform.

I'd just rather use network file storage for document sharing. I'm sure there are plenty of companies that don't want to give Microsoft all of their documents.

Network file storage has several big disadvantages compared to O365:

1. Versioning
2. Retention policies
3. DLP
4. Other compliance policies (such as sensitivity ratings)

Yes, I'm sure you can add products on top of a file server to accomplish these things (and yes, I know there are file systems which support versioning), but M365 not only has them for SharePoint, but for things like retention, DLP, compliance, etc, the policies can apply to ALL workloads. These are the main reasons clients ask us to get rid of their file shares. I'm doing an engagement with a client in Asia in a few weeks where the entire goal is eliminating their file servers. We're getting more and more of these specific engagements.

 

[Scarpozzi]

I may have given too much notice. I got slammed with a ton of requests for stuff the past few weeks as people are going into panic mode about me leaving. I just have 4 days after today and then I'm starting the new responsibilities and dropping my old ones.

I used to have a really nice office when I worked in IT. The office had a window, but also had windows that were open to the hallway (building design) that allowed anyone passing to look in on you. My new office was a conference room, so picture a long corner office with 2 huge windows...only a solid door with no windows to the hallway (or windows on the door either). The guy I'm replacing had a different office, so this is definitely a step up from where he was.

I've got a decent idea about the job responsibilities and it appears that he actually delegated most of the busy work of the job to the guy that reports to me. Oddly enough, I don't like that....I'm going to see if I can get a handle on what all is being done and see if we can programatically solve the work riddles to make his job a lot easier (and make him easier to replace if he happens to jump ship).

I think things are going to be alright and I may have time to continue neffing here afterall!

 

[Red Squirrel]

Single point of failure? How do you figure? Do you seriously think your company has more redundancy than Microsoft, Amazon, or Google? MS replicates the environments between datacenters for service redundancy and has an incredible amount of redundant bandwidth.

Your costs to match the redundancy of a Microsoft or Google far exceeds their current licensing costs. You'd have to pay the salaries of highly skilled people to design, build, and maintain it. You'd have to pay for a lot of expensive hardware, software, and bandwidth. You'd have to pay licenses for a lot of additional software as well. The fact is, if technology is not your company's core competency, there is no way you can achieve what Azure, AWS, and Google Cloud can achieve at their price points.

The provider could always have some very serious failure at their end, or decide to make some drastic change that affects your operation. Or in the case of free services, they can just pull the plug. While those scenarios are not super likely, they're still there and they are out of your control.

Suppose the same can be said for having bare metal hosted in a data centre though, the compnay could go tits up and shut down all the servers without notice. Either way, having proper backups that are on premise is very critical. And if your data is on prem then having off site backups.

One thing I need to read up more is containers, they seem to be all the rage now. If I understand it correctly you should be able to setup an environment and easily transplant it to another server. This makes backups and disaster recovery easier.

 

[Scarpozzi]

One thing I need to read up more is containers, they seem to be all the rage now. If I understand it correctly you should be able to setup an environment and easily transplant it to another server. This makes backups and disaster recovery easier.

I was doing this stuff 20+ years ago...it's nothing new...but there are lots of different layers you can consider containers.

When working with linux, I would always setup servers as generic boxes not doing too much to make them special on VMWare. If they needed SAN space for something big, I would configure a standard EXT2 or EXT3 volume (depending on journaling needs of the file system), then mount it to the VMware guest OS and use VMotion if needed to have it switch VMware host nodes. I didn't bother doing much with VFS when I was working with finite constraints of the SAN and didn't want to deal with that abstraction layer and its nonsense. I would just size things appropriately on the front end of its service life and mount additional volumes if needed and migrate data appropriately. I was also selecting with virtual hard disks and other things that added additional technologies in the mix, in case we ever needed more direct disk access. I was actually able to mount those volumes directly to the tape library from a guest OS and do direct restores from tape without the network latency we got using the web clients. Those are things to consider when doing large recoveries.

VMware and other systems can do a lot with snapshots. Quite a few NAS/SAN technologies can do the same thing with their file systems too. As long as you have the software, you can do all kinds of things with containers. I also, however, used multiple service IPs to totally revolutionize how we handled listeners. I would limit which ports could talk to IPs on both sides of our load balancers and would force almost all traffic to the data center through the reverse proxies. This allowed us to firewall everything twice and it was really easy to identify hackers and other traffic that were attempting to connect on any other ports than the 1-10 we used per IP. (including passive ports)

Anyhow...I'm way out of touch on what people are doing these days, but could jump in pretty easily. The core is likely the same as always, just being rebranded and tweaked before being sold to technology managers as the latest/greatest tech.

 

[Red Squirrel]

Oh yeah VMs are nothing new, I was talking about actual containers done in Docker, they're a new thing that has been all the rage in the past few years. it's sorta like a VM but not really a VM. I don't quite get the use case though, I guess they have less overhead since you're not running a full blown OS for each one.

Been meaning to update my actual VM infrastructure too, I want to have at least 2 hosts and do HA and stuff. Proxmox is the way to go for that for home use. Vmware is limited without shelling out big bucks and think they charge per VM now too.

 

[IndyColtsFan]

The provider could always have some very serious failure at their end, or decide to make some drastic change that affects your operation. Or in the case of free services, they can just pull the plug. While those scenarios are not super likely, they're still there and they are out of your control.

That's not a logical argument. Your main internet line is much more likely to be cut by a backhoe and you lose connectivity to all of your clients. Or what if there's an earthquake or hurricane and your entire datacenter is wrecked? Good luck paying for a second, redundant datacenter and bringing up all the infrastructure even if you're colocating it. And I promise you that you won't have near the redundancy of an Amazon, Google, or Microsoft especially for what you're paying them.

No offense, but I don't think you are grasping the scale I'm talking about here although even that is less and less an issue - I'm seeing companies with under 500 employees going to the cloud as much as I am seeing the expected 400,000+ user companies going to the cloud. They get it.

Suppose the same can be said for having bare metal hosted in a data centre though, the compnay could go tits up and shut down all the servers without notice. Either way, having proper backups that are on premise is very critical. And if your data is on prem then having off site backups.

One thing I need to read up more is containers, they seem to be all the rage now. If I understand it correctly you should be able to setup an environment and easily transplant it to another server. This makes backups and disaster recovery easier.

Outside of a few specific cases, there is no reason - none - for a company whose primary focus isn't technology to invest in datacenters and a huge IT department. It just doesn't make economic sense. It makes more sense to push all of the mundane and boring server work up to the cloud and free your IT resources to add actual value to the business via things like business process automation. Which company wants to have to upgrade servers, operating systems, applications, etc. on a routine basis? None. Leave it to the experts to do that.

 

[Red Squirrel]

I guess I just like being in control. It's kind of like a construction company just subcontracting all the work instead of having their own equipment and doing the work themselves. What's even the point if you're getting everyone else to do it for you?

There seems to be this odd pride these days of not actually doing anything and getting someone else to do it. Manufacturing is like that too. Take a company like Apple (but really almost every electronic company), they don't even MAKE anything. They just get another company to do it then slap their name on it and own the IP rights. This whole movement just seems so odd to me.

 

[MrSquished]

I guess I just like being in control. It's kind of like a construction company just subcontracting all the work instead of having their own equipment and doing the work themselves. What's even the point if you're getting everyone else to do it for you?

There seems to be this odd pride these days of not actually doing anything and getting someone else to do it. Manufacturing is like that too. Take a company like Apple (but really almost every electronic company), they don't even MAKE anything. They just get another company to do it then slap their name on it and own the IP rights. This whole movement just seems so odd to me.

This all makes no sense.

 

[Red Squirrel]

They are all different forms of outsourcing. What I'm getting at is, if you outsource the very bread and butter of your business, then why even be in that business in first place? Why bother running an IT company if you're just putting everything in the cloud?

 

[MrSquished]

They are all different forms of outsourcing. What I'm getting at is, if you outsource the very bread and butter of your business, then why even be in that business in first place? Why bother running an IT company if you're just putting everything in the cloud?

He said if a company's primary focus is not technology, there is definitely no reason to try to do everything in house.

Let's take your theorizing to another type of business, since you also extrapolated it to construction and subcontracting there. I'm helping a guy flesh out a potential food business. One thing we need is lots of bread. We are sourcing a bakery to make all the bread for us. You think we'll have the space to install commercial ovens to pop out hundreds of pieces of bread in expensive real estate markets in urban areas where our store fronts are not going to be that big? It's just not feasible to not outsource it. We are going to sell beverages, should we make them all in house too? I think Honest Tea and Fiji water might be better options for us there. What about our Point of Sale machines and software that tracks our sales and inventory. We are going to outsource that too. Marketing? I'll probably handle a bunch of it, but we will also be outsourcing some campaigns to marketing firms. There are a million examples where it makes more sense to outsource parts of a business to companies that specialize in that segment.

 

[ultimatebob]

I guess I just like being in control. It's kind of like a construction company just subcontracting all the work instead of having their own equipment and doing the work themselves. What's even the point if you're getting everyone else to do it for you?

There seems to be this odd pride these days of not actually doing anything and getting someone else to do it. Manufacturing is like that too. Take a company like Apple (but really almost every electronic company), they don't even MAKE anything. They just get another company to do it then slap their name on it and own the IP rights. This whole movement just seems so odd to me.

I don't know about where you work, but the company I work for is kinda cheap. If they built an on-prem data center, they would probably cut corners on things like UPS backups and generators to the point where we screwed during a lengthy power outage. Living on the Connecticut coastline, those are known the happen every few years or so.

Besides, 2/3rds of the company works remotely thanks to COVID, so most of us wouldn't even see a performance improvement if everything was hosted locally.