• Anonymous (unregistered)

    It's a WTF when there's not enough process, it's a WTF when there's too much process, when the hell are you guys going to be happy? Exactly the right amount of process? Do you think that actually exists?

  • I_M_Noman (unregistered)

    I think I used to work for that company too...

  • (cs)

    What I want to know is how did Bruce W. manage to get the project approved in the first place?

  • dangb (unregistered) in reply to Anonymous
    Anonymous:
    It's a WTF when there's not enough process, it's a WTF when there's too much process, when the hell are you guys going to be happy? Exactly the right amount of process? Do you think that actually exists?

    It's the mythical Goldilocks Zone of Process, just right.

  • Machtyn (unregistered) in reply to jo42
    jo42:
    What I want to know is how did Bruce W. manage to get the project approved in the first place?

    Obviously the New Products Proposal Approval Division has its process streamlined. They only have to randomize which of the two rubber stamps to pick up and use.

  • (cs)

    "big if a deal!"

    Is that supposed to be "big of a deal" or "big effin' deal" ? ;)

  • greg (unregistered)

    ...but he did get his servers. The process be praised!

  • Jon (unregistered)

    The trick is to work for a company with no clients, have very little process, and be aware of and ok with the fact that things will break occasionally, but the cost of them breaking is much less than the cost of dealing with the red tape on a regular basis.

  • SomeWhiteGuy (unregistered) in reply to I_M_Noman
    I_M_Noman:
    I think I used to work for that company too...

    I think I work for that company now. It's a headache to get anything approved. Luckily we do have a 'Streamlined' project that can be pre-approved and moved to production in 24 hrs.

  • Dan Meltzer (unregistered)

    Week 24: Comment!

  • Nodody (unregistered)

    I don't think this story is about bureaucracy. Yeah, there's a needlessly long form that you have to submit in the correct manner. Other than that though, the bureaucracy didn't seem to be that deep. The "process" part of the WTF ended after week two (and up to that point, it sounded like the communication was reasonably good...he got a prompt response when he sent the form to the wrong person, for example). To me, this WTF is about incompetence and people flat out not wanting to do their jobs, because that is the only possible explanation for the next 18 weeks.

  • Sean (unregistered) in reply to jo42
    jo42:
    What I want to know is how did Bruce W. manage to get the project approved in the first place?

    My general policy is "it's easier to ask for forgiveness than permission". Bruce W. probably started working on the application, deploying to the machines he had, before he even put in a project proposal. That way, when you go to the project approval meeting and your boss's boss's boss's...boss asks what the timeframe is, you can say "it's already done" (since you had the three weeks it took to get an approval meeting scheduled to work).

    Oh, sure, they'll get a little red in the face, but they won't sh**-can a project that's already done.

  • Anonymous Coward (unregistered)

    Probably a stupid question, but what's the difference between a testing and staging server?

    CAPTCHA: lupatum. A werewolf's belly?

  • jeff (unregistered)

    I've been there. I started a request for a new "database" on an existing database server. After a month of bureaucracy, papers, emails, forms, ect., I got the database created. After 2 weeks, I got my databases migrated. Then along came security; it took two weeks to get security to agree to give me user access, but then the security guy quiet, so I had to go through the process again. The whole process took so long I eventually just gave up. Ridiculous!

  • (cs)

    Week 18 - Heads start rolling...even poor Bruce's. snip Week 20 - Bruce distributes the application URL pointing to the brand new servers.

    Did you somehow forget to mention, "Week 19 - Bruce gets resurrected from the dead"?

  • (cs) in reply to Anonymous Coward
    Anonymous Coward:
    Probably a stupid question, but what's the difference between a testing and staging server?
    Simply put, a testing server is where you do testing, and a staging server is where you do staging.

    Don't tell me you didn't see that coming.

  • (cs)

    "other projects are ahead of us"

    You're breaking character.

  • jimi (unregistered)

    The real WTF is all the changes between first and third person.

  • RBoy (unregistered)

    What does your comment require? -To be first

    What is the estimate for the hardware of this comment? -$0. Using all existing equipment

    What is the estimate of labor of this comment? -20 seconds

    Will any of this comment be outsourced? -No I will be typing it myself

    What is the time frame for this comment? -Moments after the main post, and before any other comments.

    Who will be the Project Manager for this comment? -RBoy

    Who will have final approval after the comment has been completed? -Alex

    ... Comment form approved. ...

    "First"

  • CGomez (unregistered)

    This isn't that strange. We have only about 350 employees and it usually takes one year just to reach project kickoff. Then a year is spent trying to "keep costs down". Typically at this point customers have already been sold on the project for about a year and start to wonder if it's vaporware. All the while the development system hums along. Sure we may not be truly production ready, but we would only know that if we had environments to promote to and test.

    Finally the whole project is killed when the last potential customer gives up waiting.

    Usually takes about three years total.

  • (cs) in reply to SomeWhiteGuy
    SomeWhiteGuy:
    I_M_Noman:
    I think I used to work for that company too...

    I think I work for that company now. It's a headache to get anything approved.

    I work for that company too. And that kind of mess exists when you do work within one division. Last year we had a project that involved three divisions and wireless networking. It basically took over a year of bureaucracy and then vanished without a trace.

  • Steve (unregistered)

    I had much the same kind of thing happen to me. I requested 10K of SAN disk for an Oracle DB. Instead of simply purchasing the disk-space IT Services decided to design an entire new environment for us. They then gave us the bill: 60K for a project team and basically a blank cheque for them to build an environment.

    The whole thing cost the project 100K in the end and it took IT Services over a year to provide us with the new DB production servers.

  • Idhrendur (unregistered) in reply to SomeWhiteGuy
    SomeWhiteGuy:
    I_M_Noman:
    I think I used to work for that company too...

    I think I work for that company now. It's a headache to get anything approved. Luckily we do have a 'Streamlined' project that can be pre-approved and moved to production in 24 hrs.

    I'm pretty sure I'm working for that company too. Fortunately for me, I'm just a lowly software engineer, and don't have to deal with any of this mess. The poor guy in the next cubicle, however...

  • that's right (unregistered)

    basically any multi hundred million dollar corporation will have this red tape. There is a reason this bureaucracy exists, a lot of it is CYA, however, if something does go wrong they need to know WHERE it went wrong, WHY it went wrong, and (depending on what it was) WHO went wrong... so that something like that does not happen again.

    A "simple outage" (hardware conflict, driver conflict, a process not starting etc) can cost a company hundreds of thousands if not millions of dollars. Sure the red tape is expensive as well as getting approval is slow, but mistakes can be even more expensive.

  • (cs) in reply to Code Dependent
    Code Dependent:
    Week 18 - Heads start rolling...even poor Bruce's. *snip* Week 20 - Bruce distributes the application URL pointing to the brand new servers.

    Did you somehow forget to mention, "Week 19 - Bruce gets resurrected from the dead"?

    No, Bruce simply rolled along with his head.

  • (cs) in reply to powerlord
    powerlord:
    "big if a deal!"

    Is that supposed to be "big of a deal" or "big effin' deal" ? ;)

    The normal form here would be 'not that big a deal'. The 'of' (or 'if') has always seemed superfluous to me.

  • (cs)

    The Actual WTF is that the application was deployed at the 8-week mark anyway, despite the lack of an adequate production server environment for it to run on.

    The result: no consequences for the Bureaucracy's failure to meet business needs in a timely manner, and therefore no hope of anything ever changing.

    There sometimes comes a time where it's appropriate to say, "This application was released three months behind schedule because of unexpected delays introduced by Division X. Because of these delays, the company missed out on Y million dollars of revenue and caused customer Z to end their business relationship with us."

  • AdT (unregistered) in reply to Code Dependent
    Code Dependent:
    Did you somehow forget to mention, "Week 19 - Bruce gets resurrected from the dead"?

    Duh... that's just the anonymization. Bruce's real name is Lazaruce, of course.

  • (cs) in reply to Anonymous
    Anonymous:
    It's a WTF when there's not enough process, it's a WTF when there's too much process, when the hell are you guys going to be happy? Exactly the right amount of process? Do you think that actually exists?

    This sort of hardware crap is endemic to big companies, and it sucks the big one. I request servers all the time, just so when I need one, I can grab one out of storage, and even then, chances are I'll have to buy one under the table using money that isn't approved for capital purchases, because most of my requests are denied for random budget reasons. Like, I got money to upgrade the whole property to Vista (which I have no intention of doing, and didn't ask for any funding), and I got denied on my new servers. WTF?

    I ended up basically setting up my network services (DHCP as a bunch of desktops set to PXE boot off one big app server. If one dies, I just swap it out, point it at the main server, and let it go. Before we had the big nice main server (which cost all of 2500 dollars; my gaming desktop cost more than that), I ran the whole thing off of yet still another desktop.

    This is why small companies beat the crap out of big companies when you need to be able to move quickly.

  • tt (unregistered) in reply to jimi
    jimi:
    The real WTF is all the changes between first and third person.
    ...and skipping second one completely!
  • AT (unregistered)

    [to prevent cleaning staff misfeance]

    Assuming "misfeance" is intended to mean "misfeasance" which is defined as "Improper and unlawful execution of an act that in itself is lawful and proper."

    I'm trying to figure out what a cleaning staff can do to make turning off a power strip illegal...

  • (cs) in reply to Anonymous Coward
    Anonymous Coward:
    Probably a stupid question, but what's the difference between a testing and staging server?

    "Testing" is where the developers test our changes basically to make sure all of our code made it from one environment to the next (i.e. did all of the code files, images, sp's, etc get promoted). "Staging" is where the actual testers make sure the code works, meets all of the business rules, and does what it's supposed it to.

    That's just what we do here.

  • TrappedInTheMachine (unregistered)

    Sounds a bit like my present project... of course we are going on YEAR NUMBER 5 and still have not made much progress with the red tape and such... pity too, since I think it would take us about 3 to 6 months to actually do the work, test and release it.

  • (cs) in reply to Anonymous Coward

    Possibly, a staging server is one which is exactly like production and will in fact become production by, e.g., repointing the load balancer. Whereas test will never become production, and might be slightly different (e.g., machine is less powerful)

  • (cs) in reply to AT
    AT:
    [to prevent cleaning staff misfeance]

    Assuming "misfeance" is intended to mean "misfeasance" which is defined as "Improper and unlawful execution of an act that in itself is lawful and proper."

    I'm trying to figure out what a cleaning staff can do to make turning off a power strip illegal...

    Perhaps it is intended to mean "miscreance", or "miscreants" -- deliberately turning off the power strip while sniggering about their mischief.

  • Stormrider (unregistered)

    Surely the real WTF is why you need 4 WEB servers without an internet connection?

  • JustACodeMonkey (unregistered)

    We had been running one server as both our QA And production environment. But once we had apps that people were actually using we tried to get a QA server it took nearly a year to aquire a VMWare server running IIS With a total cost of $3US per year. The cost in time spent getting the server over $175K....

  • (cs)

    That picture of the ruler next to the pile of documents hit home. I documented a new system in about 30 pages (features, instructions to launch, monitor, stop and trouble shoot). The bean counters said that it wasn't "substantial enough". I "rewrote" it by including configuration and script listings in a new examples-section, added lots of screenshots, and built it up until it made a hearty thump when dropped on a desk (about 275 pages). Only then was it deemed sufficient.

    Sigh.

  • anybodyatall (unregistered)

    Isn't the WTF that he didn't escalate the timing problem to his boss and so on up the chain ?

  • letmegooglethatforyou.com (unregistered) in reply to AT
    AT:
    [to prevent cleaning staff misfeance]

    Assuming "misfeance" is intended to mean "misfeasance" which is defined as "Improper and unlawful execution of an act that in itself is lawful and proper."

    I'm trying to figure out what a cleaning staff can do to make turning off a power strip illegal...

    u're killin me.

    mis⋅fea⋅sance   /mɪsˈfizəns/ –noun Law.

    1. a wrong, actual or alleged, arising from or consisting of affirmative action.
    2. the wrongful performance of a normally lawful act; the wrongful and injurious exercise of lawful authority.
  • Joshua (unregistered)

    I find that the amount of this type of bureaucracy dramatically increases in Microsoft infrastructures. This is because if you sneeze the wrong way at the wrong server, you have to pay the Microsoft Sneeze-License fee... and if you don't, and you're audited, you have to pay three times the value of the highest valued commercial Sneeze License. Seriously - what's the deal with the stupid licensing on SharePoint and Enterprise CAL's when you can get the same functionality out of Plone for FREE?

    I can request Linux VM's all day for open source based projects with a few hours turnaround... but if it's Windows and IIS, I am required to justify the expense, and then the Server Group will figure out "when their budget allows" for the license fee for the new server. Even if it's for a simple one-off web app.

    It's stupid how paranoid companies are about angering Microsoft... and that paranoia is a big source of this kind of stuff.

  • Fedaykin (unregistered) in reply to Anonymous Coward

    Testing server: application runs in an environment that looks at development resources (db, etc.) and is not exposed to real users. Used to develop and test an application.

    Staging server: application runs in an environment that looks at production resources (db, etc.) but is not exposed to real users. Used to verify the application really works against production resources.

  • Robert S. Robbins (unregistered)

    I have put in a Service Request for a comment on The Daily WTF. It should appear in six weeks.

  • x (unregistered) in reply to that's right
    that's right:
    basically any multi hundred million dollar corporation will have this red tape. There is a reason this bureaucracy exists, a lot of it is CYA, however, if something does go wrong they need to know WHERE it went wrong, WHY it went wrong, and (depending on what it was) WHO went wrong... so that something like that does not happen again.

    A "simple outage" (hardware conflict, driver conflict, a process not starting etc) can cost a company hundreds of thousands if not millions of dollars. Sure the red tape is expensive as well as getting approval is slow, but mistakes can be even more expensive.

    depending on the company, maybe, but such companies don't allow "outages" if they have a half way intelligent design team. ala google.

    for the rest of the world, half the IT infrastructure or more was probably created specifically for the red tape to be properly implemented. in such a case one might argue and outage is a boom to productivity as all of a sudden, the tape dispenser is empty and people are....shudder....forced to think....

  • (cs) in reply to Anonymous Coward

    At my last gig, a staging server is used to test the implementation method. So you test the code on the testing server, then create a list of scripts to run and files to install for implementation, and that is tested by staging.

  • Bruce W (unregistered) in reply to Stormrider
    Stormrider:
    Surely the real WTF is why you need 4 WEB servers without an internet connection?

    I knew I forgot something! Crap, now I have to start a new iteration of PAIN.

  • Bruce W (unregistered) in reply to anybodyatall
    anybodyatall:
    Isn't the WTF that he didn't escalate the timing problem to his boss and so on up the chain ?

    My boss and my boss' boss knew all about my troubles. Yes, we didn't inform the uber-powerful person way up until later, but we shouldn't have needed to.

  • BOOM! (unregistered) in reply to x
    x:
    that's right:
    basically any multi hundred million dollar corporation will have this red tape. There is a reason this bureaucracy exists, a lot of it is CYA, however, if something does go wrong they need to know WHERE it went wrong, WHY it went wrong, and (depending on what it was) WHO went wrong... so that something like that does not happen again.

    A "simple outage" (hardware conflict, driver conflict, a process not starting etc) can cost a company hundreds of thousands if not millions of dollars. Sure the red tape is expensive as well as getting approval is slow, but mistakes can be even more expensive.

    depending on the company, maybe, but such companies don't allow "outages" if they have a half way intelligent design team. ala google.

    for the rest of the world, half the IT infrastructure or more was probably created specifically for the red tape to be properly implemented. in such a case one might argue and outage is a boom to productivity as all of a sudden, the tape dispenser is empty and people are....shudder....forced to think....

    i think the word u're looking for is boon.

  • Dazed (unregistered)

    20 weeks? Not too bad. I've just come off a project where, 10 months after the start of the project, 8 months after they found someone who knew which forms had to be filled in, 6 months after all forms had been filled in to the bureaucrats´ satisfaction and 3 months after the system was supposed into production, the IT department was saying that they still couldn't give an estimate for when they would start work on our servers ...

  • Vlad Patryshev (unregistered)

    Yeah... and imagine it is your 20% project at Google. And your expected QPS is about 90. And your script you want to deploy is 7.5k zipped. And for 3 years you hear that there are no servers capacity to deploy and serve 7.5k 90 times a second. And you have to follow the rules that change while you follow them. And it is all your fault.

Leave a comment on “The Mega Bureaucracy”

Log In or post as a guest

Replying to comment #241498:

« Return to Article