One Way to Get IT's Attention | Hacker News Reader

One Way to Get IT's Attention

omninerd.com

19 pointstom6a21 comments

At my current job (very large corporation), I requested certain specs for my development machine, basically just lots of RAM and an SSD. I said I would be more than happy to pay for the hardware myself and they could have IT install the standard desktop.

It took 6 weeks for this request to be emailed around to various people for approval and no one (I saw the email chain) had the balls to make a decision or even give their opinion on the matter. Luckily, it finally hit someone that had both a brain and balls, whose reaction was basically "of course, give it to him immediately, the extra cost is meaningless."

The point of this is, at least to some degree, costs are not what hampers IT, you can offer to pay for something and they will still not let you have it. At mt previous job I got some very suspicious looks from a manager for bringing my own printer to work (to save 5 minutes a day walking to the printer).

I've seen it in many companies.....there's something else going on other than just budget, but I can't put my finger on it.

For many people, including the lower ranks of large IT departments, it simply isn't part of their job to ever make a decision or use their own opinion.

If you think for yourself and get it wrong then your boss won't be happy. If you just pass the request around then you can't be blamed.

When people wonder why big companies make stupid decisions and why they are are so slow and inefficient... this is it.

If an entrepreneur sticks his neck out for an opportunity that has high risk but potentially high rewards, he's doing it because he expects to get a large portion of the rewards if he's successful. In a big "enterprise" company, your share of the reward for sticking your neck out is likely to be, if you're lucky, nothing more than a Lucite plaque and a mention in your review. But your share of the risk could be quite a bit larger (loss of your job for example). All the individual incentives in most organizations are tilted toward risk avoidance.

Politics are the behind-the-scenes battles that permeate the atmosphere but can't be seen. It cows people from taking responsibility and ownership.

This is the battle I face every day. I am a field tech at a rather large corporation, and there are so many departments in IT that no single one of them can do anything without stepping on another's shoes. I try to be a peacemaker, which has led to some improvement, but politics definitely hampers efficiency, and it's one reason I need to not work for big companies.

I agree with most of your post and there's probably context I'm not aware of but bringing a printer to work to save 5 minutes (and more importantly: a walk) ?! A walk is good for you!

I've been on both sides of this, and I can say with some certainty that the two biggest reasons for this attitude about hardware are 1) difficulty in managing a heterogeneous computing environment, and 2) the appearance of unfairness.

Here's another question: why is it that execs tend to always get the snazziest new gear, not the people doing real computing? Argh.

It's possible it is a capital cost vs expenses thing.

Where I worked it was very hard to get a new machine because it was a capital expenditure, which affected tax, depreciation and budgeting. Ongoing expenses were easy though, because the tax effect was different.

He mentioned billable rate, which might be the reason this isn't going to get fixed. If you're billing a client $250/hr for your employees' time, your incentive as a company is to maximize the time it takes to do things. Adding random 3 minute pauses to the workflow of your most expensive people would be an example of a "good thing" from a strictly financial perspective.

But then companies are seldom that aware of what's going on. Back in my Consulting Engineer days, I spent plenty of time trying to explain the simple math: "since you're billing me out at a 3.5 multiplier, it actually makes you more money to give me a raise". It never flew. So eventually I did.

Personally, I'm all in favor of big companies behaving this way. Less efficiency in other shops is a competitive advantage for mine.

What was the lag situation after an OS reimaging or reinstall?

Take a snapshot of a fresh OS install, with all necessary applications installed, running on the old hardware. Compare that with a snapshot of the old OS (before reinstallation), running on the new hardware. Generally, unless the hardware is really old, reinstalling Windows will get you a snappier system back.

If you're using an old HDD, consider swapping for a SSD.

If a little more performance is needed, consider overclocking the CPU by 15-20% if that's possible (a cpu cooler is much cheaper than a new machine).

Additionally, you should disable all graphic "special effects", Aero and other crap that comes with Windows.

I don't see overclocking being approved without incredibly tactful speaking

If there's an IT department to manage the PCs then he shouldn't have to do any of this. "The computer is inadequate for my work" should mean that IT come out, establish what they can do and make it happen.

If configuration changes like Aero or an SSD make a decent difference then the IT department are best placed to make them. They know all of the possible side effects, and when they get it right the benefit will reach everyone in the company.

Be careful, because sometimes disabling Aero actually makes the system slower. With Aero enabled, rendering the desktop and window UI/compositing/etc is done on the GPU. With it disabled, all that UI is now contending for CPU time.

I've had a lot more success with this tactic in small businesses than I have with large. It's really easy to go to an owner operator and express PC hardware costs relative to employee salary. For a $35k/year employee, an $875 PC upgrade is only 2.5% of their salary. It's hard to argue that you won't get a 2.5% productivity increase from keeping users in new PCs every couple of years. It's amazing how much more willing someone is to do their work when you make it less frustrating for them.

There's another obstacle besides IT: co-workers.

I once worked with a guy who used a 10 year old blurry "CRT" monitor. Repeatedly, I told to at least switch to LCD, but he wouldn't do it, citing that it would require a "business justification". If I were him, that monitor would have met with an unfortunate accident.

In such an environment, asking for a workstation that goes beyond "secretary grade" specs paints one as a gear-dandy.

In one of my first jobs as a coder we had little crappy PCs for development. They were ok, but a bit slow. When I left I wrote a polite email to the CEO pointing out the same thing - lag time x wages per hour x number of people in the team. A week later one of the devs emailed me and said they'd just recieved amazing new machines.

It's not always clear that faster machines really produce net productivity gains, unless the employee's time is already saturated with work. The net gains may manifest themselves into lower working hours, but the company can't just then start paying employees for 7.5 hours instead of 8 (although it can let go of 5% of the workforce to the same effect).

But employees can get more done in their 7.5 hours then they could in 8.

Rather, those 7.5 hours are less interrupted by waits, lag, pauses and compiles. Switching from IDE to reference is more seamless, and so the preservation of flow is much more likely.

I don't know what makes you think that you've achieved anything. Or, in other words, your main problem remains unsolved. :)

" but that has allowed me to experience a migration from Windows 2000 to Windows Vista."

Well there's your first mistake. This could just as easily be an argument to stop using crappy MS products.

(yhea yhea, i know, never going to happen).

"He urged me to summarize my data and findings in an email and then he would pass it on to some VP in the IT world."

ha. mail > /dev/managment_blackhole

Reply on news.ycombinator.com