Walzmyn writes:
"The company I work for is not a tech company. We are, however, a multi-national, multi-billion dollar company that claims to be the largest of our kind in three industries (and second largest in a 4th). And yet, our company network sucks. There is a mishmash of Citrix and SAP, multiple web-portals, and none of them work with each other. The several thousand non-technical people that work for this company are routinely asked to interface with this system and end up spending time with the helpdesk or with a supervisor looking over the shoulder for something that was supposed to be private.
I've heard of similar situations with other companies, so I wanted to ask the folks that live and breathe the tech sector this: Why can't a company this size get something so fundamental done right? Why can't they at least hire a third party to do it right for them?"
(Score: -1, Troll) by Anonymous Coward on Wednesday February 19 2014, @10:36PM
Why can't I dislodge my penis from your anus?
(Score: 5, Insightful) by GeminiDomino on Wednesday February 19 2014, @10:37PM
Take heart, me brother. IME, most tech companies can't do tech well, either.
"We've been attacked by the intelligent, educated segment of our culture"
(Score: 5, Insightful) by drac on Wednesday February 19 2014, @10:39PM
Everything IT is sometimes (not always!) seen as a cost centre. This is especially true for corporations where the main line of business is not directly related to, as in your example, networking. So they would do whatever is needed to get the job done at that point, and nothing more. Organizations of a certain size also have to deal with tremendous amounts of inertia and cannot simply write off previous investments to harmonize their networking infrastructure (how would they cost justify this? It's a potentially huge investment that doesn't necessarily contribute to the bottom line immediately).
Problems of this nature are hardly limited to non-technical corporations though - I've worked with truly arcane and horrifying set ups in major telcos.
(Score: 1, Informative) by Anonymous Coward on Wednesday February 19 2014, @10:54PM
This x1000. IT for most non-technical(and even some technical) companies are seen as a drain on the resources of the business.
(Score: 4, Insightful) by ArmoredDragon on Wednesday February 19 2014, @10:41PM
It sounds like the company you work for is huge. That said, they probably won't go around changing it unless there's a particular business need. "I don't quite like it" isn't good enough, it has to be an actual measurable hindrance on business operations. Changing things like this is expensive and difficult, and can even slow down the business while the transition is happening.
If you believe it is a hindrance, and you can prove that making a change to it will allow the company be in a better competitive position as a result of changing it, then go ahead and write up a report as such and submit it to a senior manager or executive, be sure to include ample proof and an executive summary for those who aren't tech literate (most aren't, and they don't necessarily have to be in order to do their job.)
If you get approved, go create an RFP and submit it to every consultant you know of and see if they can propose something much better, and get a project started on transitioning to it.
(Score: 3, Interesting) by c0lo on Thursday February 20 2014, @07:02AM
Oh, what's the point, it's too late already, will never happen... Proof in 3 steps:
(Score: 5, Interesting) by Anonymous Coward on Wednesday February 19 2014, @11:06PM
A largish university I used to work for had a very well run and competent computer services department. That department was run by people who knew computers, and who took seriously the advice they got from the people who worked there. They knew their systems well, in part because they built and/or assembled a lot of them. They did good work and were respected.
Then, faddish managers got control of the place, and started outsourcing more and more of the services. Responsibilities were poached away from computer services to other departments. Experienced staff either left or were let go, and buzzword-compliant, borderline incompetents replaced them.
Now, the old department is more or less gone, and has been replaced with a mish-mash of services that are either outsourced or scattered throughout the wider organization.
Put simply, the computer services department lost its autonomy and was broken apart. And the organization went from having a well-managed, centralized computing department to having poorly-managed, broken-up computing services.
Perhaps that is why the submitter's company is in such a bad way: the computing department lost an office-politics battle years ago, and what's left is run by non-specialists.
(Score: 5, Interesting) by evilviper on Wednesday February 19 2014, @11:20PM
Tech companies don't do any better, they just tend to be smaller, younger, and don't have that obscene amount of legacy.
I've worked with big legacy systems. I've been involved with porting, modifications and modernizations of those systems. While they can look incredibly simple on the surface, they may have hundreds of man-years of obscure legal, accounting, logistics, or other rules embedded in them. They may have their tentacles into 100 other systems for various bits of work... And some of those systems may be OS/2 boxes controlling $100,000 of building automation equipment, for multiple warehouses. And ALL of that has to be reverse-engineered from the existing application, before one new bit of code can be written. And in the mean time, the existing app still needs those routine updates to handle... whatever.
In those situations, I've reluctantly accepted the reality that it is overwhelmingly easier to put whatever hacks in-place to get the old code running largely unmodified on the new systems. Be it tablets using Citrix to connect to a server, running a 500-seat license of an old terminal emulator program, that is the only one still interpreting an ancient character set used by an old mainframe build by a company that hasn't existed in decades, now running on expensive but nearly-compatible reproduction hardware, which still needs a couple layers of legacy compatibility/emulation to run your app.
But in your case, it may just be several different departments buying different COTS solutions, which minimally do what each one needed, but are fundamentally incompatible with the others, and provide no useful form of interconnection. No matter the industry, companies have their own little fiefdoms that won't ever cooperate, even with a heavy push from senior management.
Do YOU see ALL home-page stories?
dev.soylentnews.org/search.pl?tid=1
github.com/SoylentNews/slashcode/issues/78
(Score: 5, Insightful) by WildWombat on Wednesday February 19 2014, @11:47PM
There are a ton of different reasons that IT is often so poorly done. I think the main one is that IT isn't looked at as being vital, as being a core part of what a company needs to do well in order to succeed. This leads to it being viewed as nothing but a cost center to the people who make the spending decisions.
To do IT well I think you need to following to be true:
* The leadership of the company needs to value IT. They don't just need to be willing to spend money, they need to understand the role it plays in their organization.
* They also have to be willing to spend the money.
* The CTO needs to be competent, not the CEO's cousin's son but someone who Knows Their Shit.
* It needs to be in house to the largest degree possible. If you need to buy generic software like an office suite or your OS from other companies, sure, but if an important part of what your company does is tied to some proprietary software package you're probably not going to be able to customize things and streamline them as much as needed. If you based your software on open products or just rolled your own you have the opportunity to streamline and make things more efficient.
For example: Restaurants. You look at a restaurant and you don't think "IT", you think chefs. But IT is vital to restaurants. I've worked at some in the past so I'll share some specifics. Next time you go watch how much time your server spends at one of the POS terminals entering orders and dealing with payments. Its a pretty significant fraction of the time. Those systems are pretty much universally wretched lowest bid turnkey solutions that someone did the bare minimum amount of work to customize for that restaurant. Better POS systems could significantly decrease the number of mistakes made and decrease the amount of time it takes to enter the orders, but since the people in charge don't value IT and don't understand how it can be used to make things more efficient, it doesn't happen. Even if management sees a problem or an opportunity they often can't do anything about it because the POS software, which is pretty much a core part of what they do (taking orders, processing payments), isn't in house. Its all off the shelf and they don't control it. Oh, and have you ever been at a busy restaurant when their POS system crashes? Pandemonium and lots of pissed off customers but the system still isn't seen 'core'.
I think that is why IT is almost universally done so poorly, its just undervalued. I think that things will improve in the long term as companies who do it well succeed and the ones that don't fall by the wayside. Just my thoughts.
Cheers,
-WW
(Score: 3, Insightful) by _0111000001100100 on Wednesday February 19 2014, @11:57PM
I feel like there are a lot of typical reasons for this:
They don't see or experience a problem.
Non tech oriented workers who don't understand what they are doing. (No end user training)
End users not following best practices and defined processes. (Being lazy and making one guy do all the work)
Fear of change.
Fear or misunderstandings of technology at the C level. ("Uhhh computers are bad mmmk")
Not the right software for you needs, maybe needs customization.
"How many zeroes to fully integrate my systems? That's going to take to long!"
They haven't had the right person come in and sell them a nice jucy T-bone steak with out havng to sticking their head up the bulls ass. (You know, finding that good ol' boy)
Your in house team knows what's best for the company and decides to reinvent the wheel.
The list could go on I'm sure. I say you could find the guy with the decision making power, become his friend and sell him on your utopian vision.
(Score: 5, Insightful) by takochan on Thursday February 20 2014, @01:20AM
Because they don't want to pay for it...
(Score: 3, Informative) by brocksampson on Thursday February 20 2014, @02:20AM
I have worked at universities in North America and Europe, large and small, private and public, obscenely rich and threadbare. In every instance in which the department takes control over their own IT, we get excellent infrastructure. Since, say, the chemistry department doesn't need to interface with the philosophy department, or even another university, a skeleton crew of competent hackers can keep things running smoothly. At my current university, in which IT is centralized (because the university board things we're a corporation), our "work stations" are just now being updated to Windows 7 from XP. For the vast majority of us who have been running other, more modern operating systems than XP, they now offer a Citrix interface instead of using open protocols. Thus, with the upgrade, I've lost Webdav access to our network storage; use the Citrix client, they say.
At the universities with departmental level IT management, we are still forced to interface with the mighty bureaucracies that fund our research. Many American funding agencies demand that we use absolutely ancient, proprietary software to submit proposals... we're talking stuff that was written to replace the old snail mail system and then never updated. It is a huge time sync and many departments have full-time staff whose job it is to interface with external bureaucracies because you have to be an expert in this terrible software to do anything. The EU, for whatever reason, is much smarter and migrated to web platforms ages ago. But even that is complicated enough that they offer courses in Brussels on how to use the latest version (they're always rolling out new versions, probably to justify their budgets).
Point being, it has nothing to do with being a tech company or a university; the larger the scope of the IT infrastructure, the more inertia it has and the more difficult and expensive it is to make minor changes. As a tech savvy end user it is infuriating to witness and to be forced to interact with it. But I'm also sympathetic. We have some software running in the lab that I wrote 10 years ago. Everyone is terrified to touch it because hours spent fixing it are hours not spent doing research. Thus, we maintain outdated versions of software (in virtual machines) basically out of fear and laziness.
(Score: 3, Interesting) by lubricus on Thursday February 20 2014, @03:33AM
Not trolling, really, and sorry US-centric:
Funny how we are understanding and thoughtful when it comes to balkanization and messes when comes to corporations (or perhaps just our own experiences), but when it comes to the Obamacare website or the new VA computer system, we were so quick to disparage and trivialize.
Or maybe it's just a different crowd here!
Federal systems suffer from all the above and more.
... sorry about the typos
(Score: 2, Interesting) by shortscreen on Thursday February 20 2014, @05:12AM
I work in a mostly non-tech role at a non-tech company. There's plenty for me to rant about. Having to wait 10 minutes for some shitty web app to load just so I can fill out daily checklists and whatnot. Having to shut off machinery and manipulate parts of it with brute force, because the control software accidentally jammed one moving part against another and caused the operator's GUI to get stuck in a reset loop. Parts of the corporate website that only work in a certain version of a certain browser. Random crashes and nonsense error messages everywhere. etc.
I think most non-techies just have low expectations and don't even dream of a fantasy world where everything is quick, consistent, and sensible.
(Score: 0) by Anonymous Coward on Thursday February 20 2014, @05:26AM
Almost anything managed by a bureaucracy and not directly linked to success or failure of the company sooner or later gets a mess, and face it, your IT department is to a large extent a bureaucracy as well.
(Score: 2, Interesting) by Geezer on Thursday February 20 2014, @07:26AM
A lot of us here are professional engineers. With that goes discipline and due diligence. I can't in good conscience let my inner little-boy-in-the-toy-store drive my proposals and recommendations. No good engineer does.
Sure, we've all got horror stories of epic failures in IT deployments, and it's always fun to poke sharp objects at the MBA's and CPA's.
But let's face it, we live in a real world of business. I'd love to upgrade my systems every time a new shiny comes along because it's new "the state of the art" or "current best practice." Who wouldn't? The thing is, most businesses can not, or some will not, spend money on Cadillac IT infrastructure and maintenance.
On the other hand, we often get garbage rammed down our throats due to cost considerations or idiot decision-makers. Either way, we can stroll into the bar after work with a clear conscience and crack jokes about it all night.
The happy medium is all you can hope for at most mid-sized companies, especially manufacturers where I work.
Sometimes good enough has to be good enough.
Scruting the inscrutable for over 50 years.
(Score: 0) by Anonymous Coward on Thursday February 20 2014, @04:05PM
Exactly. The truth is sometimes an organization has an IT department because they have to, not because it gives them a competitive advantage. If they didn't have an IT department, they wouldn't go very far. As long as the side of the company that actually makes money is executing, they aren't going to really notice until dysfunction is costing them business.
(Score: 1) by Walzmyn on Thursday February 20 2014, @07:36PM
That's a very good point. In this care I'd define well as: Works 90% of the time.
And I'd call freaking spectacular as : separate parts working together.
A healthy dose of common sense would be removing lots and lots of duplication and letting single pieces of software or at least single vendor solutions do what is now being done by dozens. No more than I interact with this system (I avoid it like the plague) I have 3 different username/passwords I have to keep up with for 3 different systems.
(Score: 3, Informative) by pacov on Thursday February 20 2014, @09:22AM
I work at a fairly large (130k+ employees) non-tech company. The single most important objective is production, and each site manager (120+ sites) is responsible for maintaining that flow of goods 24/7 very close to 365. Those site managers individually have the final say over implementing IT projects that have the possibility of interrupting production at their site. Something as helpful and innocuous as upgrading their DB servers will be denied, even after all testing of their applications has shown they will gain many man hours of time with the upgrades, due to a complete downtime requirement of 24 hours.
I can't tell if they don't understand the math, or if that single blip of no production instills a fear that the corporate office will fire them. Which ever it is, the spice must flow...
(Score: 2) by jcd on Thursday February 20 2014, @07:27PM
...at least in the education industry. Almost all of the other things on here are true too: tech is seen as a drain, bad management, politics and licensing &c. But the worst of what I've seen comes together over time as users need more functionality. "Oh, I need $application to do $task." District slaps it on there, and it sorta works well enough to get the job done. Worry about integration if a large number of people start using it. But then when it gets to that point, integration of $application is extremely difficult, expensive, time consuming, or otherwise undesireable, and that's how we end up with ten logins for eight products for one bloody job.
Don't get me wrong, I'm not blaming IT. I guess the blame lies on the people who don't foresee the needs of the network well enough and end up rushing to add things rather than reasoning out the best way to go about it.
"What good's an honest soldier if he can be ordered to behave like a terrorist?"
(Score: 1) by dave562 on Thursday February 20 2014, @07:31PM
I work for a company that does tech. We provide SaaS solutions to the Fortune 50 and the Federal government. We do not have the struggle of being viewed as a cost center as many IT shops do because we generate revenue. We are in a highly competitive marketplace, so we have to stay up to date on the current trends, and in many situations, have to develop our own apps to fill specific needs.
Our biggest challenge is finding qualified people, and despite the common refrain of "Pay more!" that is not the issue. We have great compensation packages. Our HR team does yearly salary reviews to make sure that our compensation stays competitive and often times adjusts them upwards because the company understands that in the long run it is less expensive to retain talent than it is to train new employees.
Despite all of that, we still have a hard time finding qualified candidates. We had a storage administrator position open for close to a year, and the guy we found is average at best. We hired him directly from EMC. The people out there in the market were even worse. It takes about a year to break in a new developer. By that I mean, getting them up to speed on secure coding practices, SDLC practices (change management, dev vs prod, UAT, etc)
The really good IT people are very, very expensive because their skills are in such high demand that they can command ridiculous salaries. Most of them are in consulting because that is where the money is. Often times the "best case" scenario is to have talented consultants deploy systems, and then pray to God that the transition to the in house guys works out (it rarely does).
It is hard to do tech well. It is even harder to do it at scale. While a corporation might have a decent team that runs the primary LOB application (SAP, PeopleSoft, whatever), their help desk might suck. Or they might have decent administrators who can keep Exchange running well, but they know next to nothing about Oracle and SQL Server.
The final challenge comes in with retaining good people. As hard as it is to do IT well, it is possible to execute individual projects successfully. The people who are good at bringing systems online, or tuning poor running systems, are not going to want to deal with the boredom that comes from running those systems day in, day out. I spent 10 years consulting because I never ran out of things to do. By the time we got one client completely dialed in, there were new clients who needed similar help. Before that, I spent over a year twiddling my thumbs at a small business (~200 employees) because I literally ran out of things to do and the company was not going to invest any more money in IT.
Now where I work, I spend a lot of time fixing other people's mistakes. I see a lot of them. The company does a lot of M&A work, so we are constantly acquiring new companies and having to integrated their systems and operations with ours. 80% of the time we just migrate their data out of whatever platform they have, standardize it with our own, and scrap the rest of the stack. On any given team of 10 IT people, you might have one guy who is a rock star, and two or three who are competent enough to be trusted on their own. The rest of the team will need close supervision and very explicitly defined tasks.
(Score: 2) by VLM on Monday February 24 2014, @03:51PM
"we have to stay up to date on the current trends, and in many situations, have to develop our own apps to fill specific needs."
Internally its all "see this cover of PC magazine, do that by next week, don't care if you've never heard about it before." And frankly it usually works out OK.
So that's how new tech is handled internally.
"we still have a hard time finding qualified candidates."
Right outta the HR job posting:
- Must have minimum 20 years experience in new technology that was just announced on PC Magazine front cover last month and/or Nobel prize and/or Turing award.
- Must have expert level experience in (insert list of keywords generated by selecting 20 random wikipedia articles, none of which are actually used on the job). Also mandatory that this list was from 2004 and has not been updated since the previous guy being replaced was hired.
- Must survive two day interview gauntlet consisting proving Fermats last theorem, inventing 3 new sorting algorithms in IBM 360 basic assembly language, implementing a quantum factoring algorithm using Radio Shack parts and an old TI-81 calculator, all for an ungodly stereotypical CRUD app dev position.
- Must be under 28 years of age, thirtysomethings are too expensive as employees and willing to relocate to outer Elbonia.
And this is how new tech is handled externally.
Spot any little difference?