I am defeated
-
@cakeis_not_alie said:
@scottalanmiller "When I think shoe string, I am thinking companies running Windows XP and consumer class routers and not having backup because "they don't feel that it is valuable.""
If they thought like that, they wouldn't be on Spiceworks asking for help making backups work even though they can't afford Unitrends/Veeam.
A lot of them are, I see that a bit. Quite a bit, actually. The IT person is trying to make backups work while the business doesn't see value in it.
And many are not on communities but are customers. We actually have customers running on budgets like that.
-
@scottalanmiller "You need to look at owner profits rather than company profits. Is there an owner taking home $5m while the company suffers? Or is the owner in neck deep with everyone? Is the owner taking risks too? Or just staff?"
I wouldn't work with a company where the owner wasn't taking risks too. Typically you don't get loans from a bank for an SMB here. You have to take loans out against the assets of hte shareholders. So I have been in situations where the shareholders have taken out loans against their houses/cars/etc in order to keep the company going. The 2009 financial crisis being one example. Half my stable had to do that.
"You have to consider lost opportunity which results in lost employment. If you are not pragmatic you might save one job today at the cost of ten jobs you didn't create tomorrow. You are viewing it as a name that you know that you want to protect, I'm looking at it as "doing the most good." And in doing the most good, hopefully not only employing the most people overall, but hopefully making every person who is already employed as protected as possible."
And I view what your'e saying as "close you heart to people you know because there is the remote possibility that if you allow them to go unemployed the magical hand of the market may create new jobs for people you don't know somewhere else. Maybe."
-
@cakeis_not_alie said:
If they thought like that, they wouldn't be on Spiceworks asking for help making backups work even though they can't afford Unitrends/Veeam.
When I think of shoestring setups, I think of companies with competent sysadmins who know EVERY SINGLE FLAW in their designs and are perfectly aware of every hole that needs patching, every system that needs updating and who are constantly trying to find a cheaper, more efficient way to do something because resources are scarce.
For me, shoestring means that everything is about** juggling priorities because there just ISN'T enough money to do it all properly, and there never will be**. So you have to make a call about what to do "right" and what you band-aid and what you roll the dice on. And you play politics and you use trial versions and you do favours for others to get hardware, or software or services that you need.
Ah, I think that you hit on a key ideological difference between us. You see there as being a "way to do things right" that has a cost associated with it. I do not. I see all IT as being in a business context. So for any business there is always enough money to do things right, if the business chooses to do so. In my way of thinking a business cannot have so little money to not do things right because what is "right" is determined by the money!
What you call a show string, I call "doing IT right". We should always juggle priorities and try to get the most from our IT investment. That's just the basics of IT. Home users can spend without regard to ROI. Business cannot. The "right" way is the way that best supports the business. Therefore, there is always the possibility of enough money to do it. It's self defining.
-
@scottalanmiller "But then, why do you consider it a shoe string budget?"
Because we're able to fit that sort of hardware into that budget only because I start the negotiations for the next refresh immediately after the last one. It's lobbying, favour trading and a lot of "not IT related chicanery" that makes getting hardware on those budgets possible. It's not off the shelf. Not all companies could do it. I manage to make those outfits an exception to the general cost rules every time.
-
@cakeis_not_alie said:
@scottalanmiller "But then, why do you consider it a shoe string budget?"
Because we're able to fit that sort of hardware into that budget only because I start the negotiations for the next refresh immediately after the last one. It's lobbying, favour trading and a lot of "not IT related chicanery" that makes getting hardware on those budgets possible. It's not off the shelf. Not all companies could do it. I manage to make those outfits an exception to the general cost rules every time.
But that's part of good IT. This is a very different situation from what you made it sound like at first. This is doing IT well, not cutting corners or on a lean budget.
-
Look at my desktop example a few pages back. That's not NTG running on a shoe string, that's simply us not wasting money because there is no reason to do so. It's just wise spending. That's all I see you describing.
I don't call "not wasting money" the same as "being on a shoestring."
-
@cakeis_not_alie said:
And I view what your'e saying as "close you heart to people you know because there is the remote possibility that if you allow them to go unemployed the magical hand of the market may create new jobs for people you don't know somewhere else. Maybe."
You say that but you overlook that part of my goal was to do the thing most likely to protect the existing jobs as well. I see anything that isn't the best thing for the people as "not the best thing for the people.* There are always risks and bad decisions, but if the intention is truly to do the best thing you can't really call that cold hearted. Going out of business because of bad spending would be the most cold hearted thing to do.
-
@Nic said:
I wonder if the non-profit world might be a good place to connect with other shoe-string IT folks. I know techsoup has some forums, although I don't know how active they are.
I work for a non-profit and helped other non-profits and I really can't relate to this rant at all.
-
@scottalanmiller said:
"You say that but you overlook that part of my goal was to do the thing most likely to protect the existing jobs as well. I see anything that isn't the best thing for the people as "not the best thing for the people.* There are always risks and bad decisions, but if the intention is truly to do the best thing you can't really call that cold hearted. Going out of business because of bad spending would be the most cold hearted thing to do."
But nowhere in my descriptions of things did I talk about "going out of business because of bad IT". It's our buddy @Dashrender there who is assuming that I'm talking about tightfisted types that won't spend on necessary IT. I am talking about companies that do the best they can with what they got, but where "the best they can with what they got" usually means breaking rules.
It probably means running unsupported configurations of some (sometimes all) things. It might mean using a Samba 3 AD instead of a Windows one or using KVM instead of VMware. It could mean trying to beat Starwind into running storage heavy, even though Anton will turn purple if you try. It can mean running production on ESXi free and even not using "the cloud" for everything, despite all the vendors and forum whores telling you that's what SMBs just need to do.
It means buying looking up at Supermicro as "nice gear" and then building your "server" out of an Asus workstation motherboard and some spent Adaptec cards you pilfered from the local university.
It means not having 4 hour enterprise support because uptime matters, so you instead buy thre of your whitebox Asus wonders, run two in HA and put he third on the shelf for spare parts.
It means a 8 node cluster of AMD Shanghai dual CPU servers in production in 2015 because "they still work" and the cluster can theoreticlaly tolerate a two node outage...and you still have enough s[pares to rebuild one node.
It means an HP laserjet 2 still in service because it's pantsing unkillable and you can get toners for the thing for a song.
Any or all of these things will get some condescending twunt on the forums telling me how that's a horrible, spectacular risk that puts everyone's jobs on the line and how that company should go out of business for not being able to afford better.
But at one time those Asus Workstation wonder Shanghai servers were new. It was expensive. And they were designed to last 10 years. With enough spares to make it there. That's making do with what you have, even if it breaks the rules. It's not putting jobs at risk.
It's making sure you can employ the maximum people with the minimum expenditure. It's making sure that the IT budget can go towards application development that streamlines some aspect of the business that frees up an entire person's worth of work, so that person can be retrained and reassigned and the company can produce a new product or enter a new market and hopefully grow and be better able to protect those jobs.
But it's protecting those jobs without obeying whitepapers. And with things being out of support. And then being OLD.
THAT is shoestring computing. At least to me. It's budgets that are tight and resources that are constrained but they are so because everyone is trying to grow the company even through a recession.
And yet, this is what I constantly see people getting crapped on for. And wild assumptions being made about the business owners being tightfisted twats, or the IT staff being incompetent because they didn't beg money that doesn't exist.
"Shoestring" is replacement cycles that take 6 years, and require planning to make those happen. They don't just appear out of the company budget, you work to make sure that money is there six years before the refresh happens. You work to evolve your IT in such a way that it can fit within that budget when the time comes, you don't go to the business and tell them how much you need.
It's a different way of thinking from how the forum regulars are used to. It's backwards, in fact. "This is what we can afford, make the most of it, because there isn't any more to give."
-
@lance "I work for a non-profit and helped other non-profits and I really can't relate to this rant at all."
Yeah, you know, the non-profit guys I work with never seem to have much in the way of problems. They get hardware and software they need pretty easily. The biggest issue they have is typically that they get a bunch of stuff they can't use/don't need and then have to reach out to their network of non=profit sysadmins to redistribute.
-
@cakeis_not_alie said:
But nowhere in my descriptions of things did I talk about "going out of business because of bad IT". It's our buddy @Dashrender there who is assuming that I'm talking about tightfisted types that won't spend on necessary IT. I am talking about companies that do the best they can with what they got, but where "the best they can with what they got" usually means breaking rules.
I guess that that is the big question. Whose rules or what rules are being broken? Can you give an example?
-
@cakeis_not_alie said:
"Shoestring" is replacement cycles that take 6 years, and require planning to make those happen. They don't just appear out of the company budget, you work to make sure that money is there six years before the refresh happens. You work to evolve your IT in such a way that it can fit within that budget when the time comes, you don't go to the business and tell them how much you need.
Wall St., in my experience, is eight year cycles. Budgeting is easier, but you get less of it. NTG does longer than six years.
-
@cakeis_not_alie said:
@lance "I work for a non-profit and helped other non-profits and I really can't relate to this rant at all."
Yeah, you know, the non-profit guys I work with never seem to have much in the way of problems. They get hardware and software they need pretty easily. The biggest issue they have is typically that they get a bunch of stuff they can't use/don't need and then have to reach out to their network of non=profit sysadmins to redistribute.
It's true we get software really cheap, but hardware is another story. I'm not going to accept hardware that costs more to maintain that it is worth, which is the instance in most cases.
-
@cakeis_not_alie said:
@lance "I work for a non-profit and helped other non-profits and I really can't relate to this rant at all."
Yeah, you know, the non-profit guys I work with never seem to have much in the way of problems.
The biggest issue I have seen in non-profit IT is the issue of having enough staff. I think almost any non-profit or for profit IT guy that you work with could relate to this.
-
@scottalanmiller "I guess that that is the big question. Whose rules or what rules are being broken? Can you give an example?"
Sure. I got into a fight with Anton just the other day about Starwind. You can read about it here.
The long story short is that I created a Starwind configuration that was unsupported. I did so for rational reasons. At the time, this was literally the only way to get what I needed to do done and meet the various requirements on the table. (Which don't need going into, nor am I going to rehash a 1.5 year old build at this juncture.)
I knew at the time of the build that Starwind didn't support storage heavy setups. Sure enough, I paid for it. The damned thing ate some LUNs. I had about 6 months of fighting with the damned thing tooth and nail before I got it stable.
It isn't running a config that is supported by Microsoft or by Starwind. I know that. I also know there aren't any real good choices for this client. Now that it's stable, i expect to see 6 years of service from this thing, but here are rules being broken in order to do what "shouldn't" be done with what's to hand.
Another good example is running ESXi free in production. You're not supposed to. There are a whole bunch of reasons. But in many cases there aren't really any good alternatives. (Though with the ecosystems around Hyper-V and KVM picking up and delivering free or low-cost solutions that is rapidly changing.)
I run hyperconverged setups on d-link switches. Vendors hate this and say it's unsupported because they haven't profiled d-link switches. I do it because I know they've profiled Dell Powerconnects and other broadcom-based switches and the Dlink stuff is the same silicon for a fifth the price. (With a tenth the features.)
Or the number of companies that - instead of a proper firewall like a Barracuda or Palo Alto Networks box - I run a Netgear WDNR3700v2 with OpenWRT and some modes I coded myself.
-
@cakeis_not_alie said:
Another good example is running ESXi free in production. You're not supposed to. .
I guess, but tons of people recommend that all of the time in SW and, probably, here but less traffic here. I see it several times a week, maybe once a day.
There are good reasons not to run it, HyperV offers a backup API, XenServer is more full features at that price - once you give up the backup API why not use XenServer?
But those are just options that should come up. If you need a free virtualization product there is nothing wrong with ESXi as long as you are cognizant of the backup implications.
-
@lance "The biggest issue I have seen in non-profit IT is the issue of having enough staff. I think almost any non-profit or for profit IT guy that you work with could relate to this."
They could, if they weren't vicious slave drivers. None of the sysadmins I know escape our tour at the helm. We all have our assigned tasks for the ecosystem. I'm a prototyping specialist, so I design the new system builds. Peter does the virtual appliance images. Nathan does the firewall configs...
I'm pretty sure the non-profit admins just sit in their lairs coordinating all their for-profit friends to extract maximum free labour from them.
But I don't mind. It keeps the clan together. And we need to share resources across the clan if we are to survive.
-
Other than pushback from vendors, are you really seeing the things that you are mentioning get pushback in the communities? These sound like normal things to me. People like @John-Nicholson and I promote cheap switch alternatives all of the time, for example. I get very little pushback on that. Maybe because I'm more forceful about why it is a good idea? Maybe I present it with a bit more "heading them off at the pass" explanation to make it hard to push back on me?
Not sure why but the things you mention are things that I say all of the time and don't get negative feedback on. Vendors, of course, would push back because they have money to be made. But I spend very little time in a situation where a vendor is going to provide me feedback and those that do quickly end up with management issues.
-
@scottalanmiller "Other than pushback from vendors, are you really seeing the things that you are mentioning get pushback in the communities? These sound like normal things to me. People like @John-Nicholson and I promote cheap switch alternatives all of the time, for example. I get very little pushback on that. Maybe because I'm more forceful about why it is a good idea? Maybe I present it with a bit more "heading them off at the pass" explanation to make it hard to push back on me?"
No, SAM, you don't get pushback because you're SAM. There's a difference between you and hoi polloi.
But yes, I get a LOT of pushback from "regular" types. John himself actually being something of a regular example. He never hesitates to tell me how I am breaking some design rule or other and that's horrible. But at least he doesn't follow it up with the personal attacks so we grudgingly get along.
Some people I can write off because they "orbit" vendors. (A bit like John orbits Anton, in fact.) They lap up anything a particular vendor has to say and regurgitate it. But there are a bunch that simply were trained to believe in one particular way of doing things and they are absolute bullies about it to anyone that doesn't believe.
One good example would be having said something to the effect of "hey, I need to ABC, XYZ and it can't use Server 2012 or Windows 8." wham 50 trolls out of nowhere telling you a magical tale about how you're a failure at everything because you aren't using the latest greatest and how this was putting your customers at a disadvantage and so forth.
Or anything networking. Cisco is apparently the answer to everything. If the answer isn't Cisco you're going straight to hell. Oh $deity I am sick of that one.
e
Most recently the cloud evangelists have been driving me nuts. Apparently anything below a 6 digit a year IT budget should just be using the cloud. We've all heard it. Just never mention privacy or data sovereignty or laws around privacy. Because then they'll outright attack you.Good times.
I don't quite know how to be much more "firm" in my conversations. You know me: I'm a hardass. I say "look, I need to do ABC but have rejected LMNOPQRSTUV as options and really don't want to go into the why of it. Let's focus on how to solve ABC without LMNOPQRSTU or V."
Inevitably it will descend into a massive shitfest of why I am doomed to die the death for not choosing one of L-V which then will collapse into a singularity as vendors for L-V then fight it out over who is best. Some of which will actually message me hate mail. (As will some of their acolytes.)
I have ended up deleting more than a few threads because of this.
Some of this is probably just nerd rage. Linux! Mac! PC! or Intel! AMD! But...isn't that what Tom's Hardware's forums are for? A place to jettison that baggage would be grand.
-
Yes, there is a ton of nerd rage, vendor follow ons and the like. That's the nature of IT, I've learned. The vast majority of the field isn't looking for great solutions or looking to learn continuously. I think all fields are like this, IT is just more vocal and social.
I think that there is no escaping that. No good way to filter it. You have to just accept that when crowd sourcing ideas you are going to get a lot of "random voices from the crowd." Name any large gathering of anonymous or semi-anonymous people where you would welcome all opinions?
